_id stringlengths 4 10 | text stringlengths 0 18.4k | title stringlengths 0 8.56k |
|---|---|---|
d2208884 | We develop a new method for visualizing and refining the invariances of learned representations. Specifically, we test for a general form of invariance, linearization, in which the action of a transformation is confined to a low-dimensional subspace. Given two reference images (typically, differing by some transformation), we synthesize a sequence of images lying on a path between them that is of minimal length in the space of the representation (a "representational geodesic"). If the transformation relating the two reference images is linearized by the representation, this sequence should follow the gradual evolution of this transformation. We use this method to assess the invariance properties of a state-of-the-art image classification network and find that geodesics generated for image pairs differing by translation, rotation, and dilation do not evolve according to their associated transformations. Our method also suggests a remedy for these failures, and following this prescription, we show that the modified representation is able to linearize a variety of geometric image transformations. | GEODESICS OF LEARNED REPRESENTATIONS |
d209439434 | It is well-known that classifiers are vulnerable to adversarial perturbations. To defend against adversarial perturbations, various certified robustness results have been derived. However, existing certified robustnesses are limited to top-1 predictions. In many real-world applications, top-k predictions are more relevant. In this work, we aim to derive certified robustness for top-k predictions. In particular, our certified robustness is based on randomized smoothing, which turns any classifier to a new classifier via adding noise to an input example. We adopt randomized smoothing because it is scalable to large-scale neural networks and applicable to any classifier. We derive a tight robustness in 2 norm for topk predictions when using randomized smoothing with Gaussian noise. We find that generalizing the certified robustness from top-1 to top-k predictions faces significant technical challenges. We also empirically evaluate our method on CI-FAR10 and ImageNet. For example, our method can obtain an ImageNet classifier with a certified top-5 accuracy of 62.8% when the 2 -norms of the adversarial perturbations are less than 0.5 (=127/255). Our code is publicly available at: https. Simultaneous confidence intervals and sample size determination for multinomial proportions. | Published as a conference paper at ICLR 2020 CERTIFIED ROBUSTNESS FOR TOP-k PREDICTIONS AGAINST ADVERSARIAL PERTURBATIONS VIA RAN- DOMIZED SMOOTHING |
d219708602 | We introduce the Generalized Energy Based Model (GEBM) for generative modelling. These models combine two trained components: a base distribution (generally an implicit model), which can learn the support of data with low intrinsic dimension in a high dimensional space; and an energy function, to refine the probability mass on the learned support. Both the energy function and base jointly constitute the final model, unlike GANs, which retain only the base distribution (the "generator"). GEBMs are trained by alternating between learning the energy and the base. We show that both training stages are well-defined: the energy is learned by maximising a generalized likelihood, and the resulting energy-based loss provides informative gradients for learning the base. Samples from the posterior on the latent space of the trained model can be obtained via MCMC, thus finding regions in this space that produce better quality samples. Empirically, the GEBM samples on image-generation tasks are of much better quality than those from the learned generator alone, indicating that all else being equal, the GEBM will outperform a GAN of the same complexity. When using normalizing flows as base measures, GEBMs succeed on density modelling tasks, returning comparable performance to direct maximum likelihood of the same networks. | Published as a conference paper at ICLR 2021 GENERALIZED ENERGY BASED MODELS |
d9482504 | Unsupervised feature learning has shown impressive results for a wide range of input modalities, in particular for object classification tasks in computer vision. Using a large amount of unlabeled data, unsupervised feature learning methods are utilized to construct high-level representations that are discriminative enough for subsequently trained supervised classification algorithms. However, it has never been quantitatively investigated yet how well unsupervised learning methods can find low-level representations for image patches without any additional supervision. In this paper we examine the performance of pure unsupervised methods on a low-level correspondence task, a problem that is central to many Computer Vision applications. We find that a special type of Restricted Boltzmann Machines (RBMs) performs comparably to hand-crafted descriptors. Additionally, a simple binarization scheme produces compact representations that perform better than several state-of-the-art descriptors. | Unsupervised Feature Learning for low-level Local Image Descriptors |
d250607914 | Randomized smoothing is sound when using infinite precision. However, we show that randomized smoothing is no longer sound for limited floating-point precision. We present a simple example where randomized smoothing certifies a radius of 1.26 around a point, even though there is an adversarial example in the distance 0.8 and show how this can be abused to give false certificates for CIFAR10. We discuss the implicit assumptions of randomized smoothing and show that they do not apply to generic image classification models whose smoothed versions are commonly certified. In order to overcome this problem, we propose a sound approach to randomized smoothing when using floating-point precision with essentially equal speed for quantized input. It yields sound certificates for image classifiers which for the ones tested so far are very similar to the unsound practice of randomized smoothing. Our only assumption is that we have access to a fair coin. adversarial examples with differential privacy. In IEEE Symposium on Security and Privacy (SP), 2019.Klas Leino, Zifan Wang, and Matt Fredrikson. Globally-robust neural networks. In ICML, 2021.Alexander Levine and Soheil Feizi. Wasserstein smoothing: Certified robustness against wasserstein adversarial attacks. In AISTATS, 2020.Alexander J Levine and Soheil Feizi. Improved, deterministic smoothing for L1 certified robustness. In ICML, 2021. . Integer-arithmetic-only certified robustness for quantized neural networks. In ICCV, 2021. | SOUND RANDOMIZED SMOOTHING IN FLOATING- POINT ARITHMETIC |
d252668796 | Deep neural networks have shown excellent prospects in speech separation tasks. However, obtaining good results while keeping a low model complexity remains challenging in real-world applications. In this paper, we provide a bio-inspired efficient encoder-decoder architecture by mimicking the brain's top-down attention, called TDANet, with decreased model complexity without sacrificing performance. The top-down attention in TDANet is extracted by the global attention (GA) module and the cascaded local attention (LA) layers. The GA module takes multi-scale acoustic features as input to extract global attention signal, which then modulates features of different scales by direct top-down connections. The LA layers use features of adjacent layers as input to extract the local attention signal, which is used to modulate the lateral input in a top-down manner. On three benchmark datasets, TDANet consistently achieved competitive separation performance to previous state-of-the-art (SOTA) methods with higher efficiency. Specifically, TDANet's multiply-accumulate operations (MACs) are only 5% of Sepformer, one of the previous SOTA models, and CPU inference time is only 10% of Sepformer. In addition, a large-size version of TDANet obtained SOTA results on three datasets, with MACs still only 10% of Sepformer and the CPU inference time only 24% of Sepformer. Our study suggests that top-down attention can be a more efficient strategy for speech separation. | Published as a conference paper at ICLR 2023 AN EFFICIENT ENCODER-DECODER ARCHITECTURE WITH TOP-DOWN ATTENTION FOR SPEECH SEPARA- TION |
d250279951 | It is essential yet challenging for future home-assistant robots to understand and manipulate diverse 3D objects in daily human environments. Towards building scalable systems that can perform diverse manipulation tasks over various 3D shapes, recent works have advocated and demonstrated promising results learning visual actionable affordance, which labels every point over the input 3D geometry with an action likelihood of accomplishing the downstream task (e.g., pushing or picking-up). However, these works only studied single-gripper manipulation tasks, yet many real-world tasks require two hands to achieve collaboratively. In this work, we propose a novel learning framework, DualAfford, to learn collaborative affordance for dual-gripper manipulation tasks. The core design of the approach is to reduce the quadratic problem for two grippers into two disentangled yet interconnected subtasks for efficient learning. Using the large-scale PartNet-Mobility and ShapeNet datasets, we set up four benchmark tasks for dual-gripper manipulation. Experiments prove the effectiveness and superiority of our method over baselines. * Equal contribution † Corresponding author; Project page: https://hyperplane-lab.github.io/DualAfford | Published as a conference paper at ICLR 2023 DUALAFFORD: LEARNING COLLABORATIVE VISUAL AFFORDANCE FOR DUAL-GRIPPER MANIPULATION |
d222133374 | Federated Learning (FL) is a method of training machine learning models on private data distributed over a large number of possibly heterogeneous clients such as mobile phones and IoT devices. In this work, we propose a new federated learning framework named HeteroFL to address heterogeneous clients equipped with very different computation and communication capabilities. Our solution can enable the training of heterogeneous local models with varying computation complexities and still produce a single global inference model. For the first time, our method challenges the underlying assumption of existing work that local models have to share the same architecture as the global model. We demonstrate several strategies to enhance FL training and conduct extensive empirical evaluations, including five computation complexity levels of three model architecture on three datasets. We show that adaptively distributing subnetworks according to clients' capabilities is both computation and communication efficient. Our code is available here.arXiv:2010.01264v3 [cs.LG] 14 Dec 2021Published as a conference paper at ICLR 2021 capabilities. However, to stably aggregate heterogeneous local models to a single global model under various heterogeneous settings is not apparent. Addressing these issues is thus a key component of our work. Our main contributions of this work are three-fold. Tal Ben-Nun and Torsten Hoefler. Demystifying parallel and distributed deep learning: An in-depth concurrency analysis. : Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018. Jonathan Frankle and Michael Carbin. The lottery ticket hypothesis: Finding sparse, trainable neural networks. arXiv preprint arXiv:1803.03635, 2018. Song Han, Huizi Mao, and William J Dally. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. arXiv preprint arXiv:1510.00149, 2015. . Improving federated learning personalization via model agnostic meta learning. arXiv preprint arXiv: | HETEROFL: COMPUTATION AND COMMUNICATION EFFICIENT FEDERATED LEARNING FOR HETEROGE- NEOUS CLIENTS |
d235795569 | Many reinforcement learning (RL) agents require a large amount of experience to solve tasks. We propose Contrastive BERT for RL (COBERL), an agent that combines a new contrastive loss and a hybrid LSTM-transformer architecture to tackle the challenge of improving data efficiency. COBERL enables efficient and robust learning from pixels across a wide variety of domains. We use bidirectional masked prediction in combination with a generalization of a recent contrastive method to learn better representations for RL, without the need of hand engineered data augmentations. We find that COBERL consistently improves data efficiency across the full Atari suite, a set of control tasks and a challenging 3D environment, and often it also increases final score performance. Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. , et al. Model-based reinforcement learning for atari. arXiv preprint arXiv:1903.00374, 2019. , et al. Human-level control through deep reinforcement learning. nature, 518(7540):529-533, 2015. | COBERL: CONTRASTIVE BERT FOR REINFORCE- MENT LEARNING |
d204806122 | To deflect adversarial attacks, a range of "certified" classifiers have been proposed. In addition to labeling an image, certified classifiers produce (when possible) a certificate guaranteeing that the input image is not an p -bounded adversarial example. We present a new attack that exploits not only the labelling function of a classifier, but also the certificate generator. The proposed method applies large perturbations that place images far from a class boundary while maintaining the imperceptibility property of adversarial examples. The proposed "Shadow Attack" causes certifiably robust networks to mislabel an image and simultaneously produce a "spoofed" certificate of robustness. * equal contribution arXiv:2003.08937v1 [cs.LG] | BREAKING CERTIFIED DEFENSES: SEMANTIC ADVER- SARIAL EXAMPLES WITH SPOOFED ROBUSTNESS CER- TIFICATES |
d252873240 | Oriented object detection emerges in many applications from aerial images to autonomous driving, while many existing detection benchmarks are annotated with horizontal bounding box only which is also less costive than fine-grained rotated box, leading to a gap between the readily available training corpus and the rising demand for oriented object detection. This paper proposes a simple yet effective oriented object detection approach called H2RBox merely using horizontal box annotation for weakly-supervised training, which closes the above gap and shows competitive performance even against those trained with rotated boxes. The cores of our method are weakly-and self-supervised learning, which predicts the angle of the object by learning the consistency of two different views. To our best knowledge, H2RBox is the first horizontal box annotation-based oriented object detector. Compared to an alternative i.e. horizontal box-supervised instance segmentation with our post adaption to oriented object detection, our approach is not susceptible to the prediction quality of mask and can perform more robustly in complex scenes containing a large number of dense objects and outliers. Experimental results show that H2RBox has significant performance and speed advantages over horizontal box-supervised instance segmentation methods, as well as lower memory requirements. While compared to rotated box-supervised oriented object detectors, our method shows very close performance and speed. The source code is available at PyTorch-based MMRotate and Jittor-based JDet. | Published as a conference paper at ICLR 2023 H2RBOX: HORIZONTAL BOX ANNOTATION IS ALL YOU NEED FOR ORIENTED OBJECT DETECTION |
d253707860 | Ensembling has proven to be a powerful technique for boosting model performance, uncertainty estimation, and robustness in supervised learning. Advances in self-supervised learning (SSL) enable leveraging large unlabeled corpora for state-of-the-art few-shot and supervised learning performance. In this paper, we explore how ensemble methods can improve recent SSL techniques by developing a framework that permits data-dependent weighted cross-entropy losses. We refrain from ensembling the representation backbone; this choice yields an efficient ensemble method that incurs a small training cost and requires no architectural changes or computational overhead to downstream evaluation. The effectiveness of our method is demonstrated with two state-of-the-art SSL methods, DINO (Caron et al., 2021) and MSN (Assran et al., 2022). Our method outperforms both in multiple evaluation metrics on ImageNet-1K, particularly in the few-shot setting. We explore several weighting schemes and find that those which increase the diversity of ensemble heads lead to better downstream evaluation results. Thorough experiments yield improved prior art baselines which our method still surpasses; e.g., our overall improvement with MSN | Published as a conference paper at ICLR 2023 WEIGHTED ENSEMBLE SELF-SUPERVISED LEARNING |
d251563977 | Linear regression is a fundamental tool for statistical analysis. This has motivated the development of linear regression methods that also satisfy differential privacy and thus guarantee that the learned model reveals little about any one data point used to construct it. However, existing differentially private solutions assume that the end user can easily specify good data bounds and hyperparameters. Both present significant practical obstacles. In this paper, we study an algorithm which uses the exponential mechanism to select a model with high Tukey depth from a collection of non-private regression models. Given n samples of d-dimensional data used to train m models, we construct an efficient analogue using an approximate Tukey depth that runs in time O(d 2 n + dm log(m)). We find that this algorithm obtains strong empirical performance in the data-rich setting with no data bounds or hyperparameter selection required. * Published as a conference paper at ICLR 2023 2022) each model is a slope, the natural notion of high depth is the median, and the user provides an interval for candidate slopes.We generalize this in two ways to obtain our algorithm, TukeyEM. The first step is to replace the median with a multidimensional analogue based on Tukey depth. Second, we adapt a technique based on propose-test-release (PTR), originally introduced by Brown et al.(2021)for private estimation of unbounded Gaussians, to construct an algorithm which does not require bounds on the domain for the overall exponential mechanism. We find that a version of TukeyEM using an approximate and efficiently computable notion of Tukey depth achieves empirical performance competitive with (and often exceeding) that of non-privately tuned baseline private linear regression algorithms, across several synthetic and real datasets. We highlight that the approximation only affects utility and efficiency; TukeyEM is still differentially private. Given an instance where TukeyEM constructs m models from n samples of d-dimensional data, the main guarantee for our algorithm is the following:Theorem 1.1. TukeyEM is (ε, δ)-DP and takes time O(d 2 n + dm log(m)).Two caveats apply. First, our use of PTR comes at the cost of an approximate (ε, δ)-DP guarantee as well as a failure probability: depending on the dataset, it is possible that the PTR step fails, and no regression model is output. Second, the algorithm technically has one hyperparameter, the number m of models trained. Our mitigation of both issues is empirical. Across several datasets, we observe that a simple heuristic about the relationship between the number of samples n and the number of features d, derived from synthetic experiments, typically suffices to ensure that the PTR step passes and specifies a high-utility choice of m. For the bulk of our experiments, the required relationship is on the order of n 1000 · d. We emphasize that this heuristic is based only on the data dimensions n and d and does not require further knowledge of the data itself.RELATED WORKLinear regression is a specific instance of the more general problem of convex optimization. Ignoring dependence on the parameter and input space diameter for brevity, DP-SGD(Bassily et al., 2014)and objective perturbation(Kifer et al., 2012)obtain the optimal O( | Published as a conference paper at ICLR 2023 EASY DIFFERENTIALLY PRIVATE LINEAR REGRESSION |
d235458091 | Multimodal contrastive learning methods like CLIP train on noisy and uncurated training datasets. This is cheaper than labeling datasets manually, and even improves out-of-distribution robustness. We show that this practice makes backdoor and poisoning attacks a significant threat. By poisoning just 0.01% of a dataset (e.g., just 300 images of the 3 million-example Conceptual Captions dataset), we can cause the model to misclassify test images by overlaying a small patch. Targeted poisoning attacks, whereby the model misclassifies a particular test input with an adversarially-desired label, are even easier requiring control of 0.0001% of the dataset (e.g., just three out of the 3 million images). Our attacks call into question whether training on noisy and uncurated Internet scrapes is desirable. | Published as a conference paper at ICLR 2022 POISONING AND BACKDOORING CONTRASTIVE LEARNING |
d226965048 | Humans can quickly associate stimuli to solve problems in novel contexts. Our novel neural network model learns state representations of facts that can be composed to perform such associative inference. To this end, we augment the LSTM model with an associative memory, dubbed Fast Weight Memory (FWM). Through differentiable operations at every step of a given input sequence, the LSTM updates and maintains compositional associations stored in the rapidly changing FWM weights. Our model is trained end-to-end by gradient descent and yields excellent performance on compositional language reasoning problems, meta-reinforcement-learning for POMDPs, and small-scale word-level language modelling. 1 | Work in progress. Under review. LEARNING ASSOCIATIVE INFERENCE USING FAST WEIGHT MEMORY |
d256231485 | Performant Convolutional Neural Network (CNN) architectures must be tailored to specific tasks in order to consider the length, resolution, and dimensionality of the input data. In this work, we tackle the need for problem-specific CNN architectures. We present the Continuous Convolutional Neural Network (CCNN): a single CNN able to process data of arbitrary resolution, dimensionality and length without any structural changes. Its key component are its continuous convolutional kernels which model long-range dependencies at every layer, and thus remove the need of current CNN architectures for task-dependent downsampling and depths. We showcase the generality of our method by using the same architecture for tasks on sequential (1D), visual (2D) and point-cloud (3D) data. Our CCNN matches and often outperforms the current state-of-the-art across all tasks considered. 1 * Equal contribution. 1 Our code is publicly available at github.com/david-knigge/ccnn. | Published as a conference paper at ICLR 2023 MODELLING LONG RANGE DEPENDENCIES IN N D: FROM TASK-SPECIFIC TO A GENERAL PURPOSE CNN |
d253080510 | Knowledge distillation is one of the primary methods of transferring knowledge from large to small models. However, it requires massive task-specific data, which may not be plausible in many real-world applications. Data augmentation methods such as representation interpolation, token replacement, or augmentation with models are applied to tackle this problem. However, these data augmentation methods either potentially cause shifts in decision boundaries (representation interpolation), are not expressive enough (token replacement), or introduce too much computational overhead (augmentation with models). To this end, we propose Aug-Pro (Augmentation with Projection), an effective and efficient data augmentation method for distillation. Our method builds on top of representation interpolation augmentation methods to maintain the diversity of expressions and converts the augmented data to tokens to avoid shifting decision boundaries. It uses simple operations that come with little computational overhead. The results on multiple GLUE tasks show that our methods can improve distillation performance by a large margin at a low time cost. Codes are available at https://github.com/ google-research/google-research/tree/master/augpro. | Published as a conference paper at ICLR 2023 AUGMENTATION WITH PROJECTION: TOWARDS AN EFFECTIVE AND EFFICIENT DATA AUGMENTATION PARADIGM FOR DISTILLATION |
d249394665 | The long runtime of high-fidelity partial differential equation (PDE) solvers makes them unsuitable for time-critical applications. We propose to accelerate PDE solvers using reduced-order modeling (ROM). Whereas prior ROM approaches reduce the dimensionality of discretized vector fields, our continuous reduced-order modeling (CROM) approach builds a low-dimensional embedding of the continuous vector fields themselves, not their discretization. We represent this reduced manifold using continuously differentiable neural fields, which may train on any and all available numerical solutions of the continuous system, even when they are obtained using diverse methods or discretizations. We validate our approach on an extensive range of PDEs with training data from voxel grids, meshes, and point clouds. Compared to prior discretization-dependent ROM methods, such as linear subspace proper orthogonal decomposition (POD) and nonlinear manifold neuralnetwork-based autoencoders, CROM features higher accuracy, lower memory consumption, dynamically adaptive resolutions, and applicability to any discretization. For equal latent space dimension, CROM exhibits 79× and 49× better accuracy, and 39× and 132× smaller memory footprint, than POD and autoencoder methods, respectively. Experiments demonstrate 109× and 89× wall-clock speedups over unreduced models on CPUs and GPUs, respectively. Videos and codes are available on the project page: https://crom-pde.github.io. | Published as a conference paper at ICLR 2023 CROM: CONTINUOUS REDUCED-ORDER MODELING OF PDES USING IMPLICIT NEURAL REPRESENTA- TIONS |
d218502175 | Neural Ordinary Differential Equations (ODEs) are elegant reinterpretations of deep networks where continuous time can replace the discrete notion of depth, ODE solvers perform forward propagation, and the adjoint method enables efficient, constant memory backpropagation. Neural ODEs are universal approximators only when they are non-autonomous, that is, the dynamics depends explicitly on time. We propose a novel family of Neural ODEs with time-varying weights, where time-dependence is non-parametric, and the smoothness of weight trajectories can be explicitly controlled to allow a tradeoff between expressiveness and efficiency. Using this enhanced expressiveness, we outperform previous Neural ODE variants in both speed and representational capacity, ultimately outperforming standard ResNet and CNN models on select image classification and video prediction tasks. | Time Dependence in Non-Autonomous Neural ODEs |
d6041587 | EXTENDED ABSTRACTThe modern incarnation of neural networks, now popularly known as Deep Learning (DL), accomplished record-breaking success in processing diverse kinds of signals -vision, audio, and text. In parallel, strong interest has ensued towards constructing a theory of DL. This paper opens up a group theory based approach, towards a theoretical understanding of DL, in particular the unsupervised variant. First we establish how a single layer of unsupervised pre-training can be explained in the light of orbit-stabilizer principle, and then we sketch how the same principle can be extended for multiple layers.We focus on two key principles that (amongst others) influenced the modern DL resurgence.(P1) Geoff Hinton summed this up as follows. "In order to do computer vision, first learn how to do computer graphics". Hinton (2007). In other words, if a network learns a good generative model of its training set, then it could use the same model for classification. | A GROUP THEORETIC PERSPECTIVE ON UNSUPER- VISED DEEP LEARNING * |
d16053675 | Triplet networks are widely used models that are characterized by good performance in classification and retrieval tasks. In this work we propose to train a triplet network by putting it as the discriminator in Generative Adversarial Nets (GANs). We make use of the good capability of representation learning of the discriminator to increase the predictive quality of the model. We evaluated our approach on Cifar10 and MNIST datasets and observed significant improvement on the classification performance using the simple k-nn method. | Workshop track -ICLR 2017 TRAINING TRIPLET NETWORKS WITH GAN |
d253117181 | We present a smoothly broken power law functional form (referred to by us as a broken neural scaling law (BNSL)) that accurately models and extrapolates the scaling behaviors of deep neural networks (i.e. how the evaluation metric of interest varies as amount of compute used for training (or inference), number of model parameters, training dataset size, model input size, number of training steps, or upstream performance varies) for various architectures and for each of various tasks within a large, diverse set of upstream and downstream tasks, in zero-shot, prompted, and fine-tuned settings. This set includes large-scale vision, | Published as a conference paper at ICLR 2023 BROKEN NEURAL SCALING LAWS |
d14704855 | Restricted Boltzmann Machines (RBMs) are general unsupervised learning devices to ascertain generative models of data distributions. RBMs are often trained using the Contrastive Divergence learning algorithm (CD), an approximation to the gradient of the data log-likelihood. A simple reconstruction error is often used to decide whether the approximation provided by the CD algorithm is good enough, though several authors(Schulz et al., 2010;Fischer & Igel, 2010)have raised doubts concerning the feasibility of this procedure. However, not many alternatives to the reconstruction error have been used in the literature. In this manuscript we investigate simple alternatives to the reconstruction error in order to detect as soon as possible the decrease in the log-likelihood during learning. | Stopping Criteria in Contrastive Divergence: Alternatives to the Reconstruction Error |
d247839131 | We consider the problem of sequential robotic manipulation of deformable objects using tools. Previous works have shown that differentiable physics simulators provide gradients to the environment state and help trajectory optimization to converge orders of magnitude faster than model-free reinforcement learning algorithms for deformable object manipulation. However, such gradient-based trajectory optimization typically requires access to the full simulator states and can only solve short-horizon, single-skill tasks due to local optima. In this work, we propose a novel framework, named DiffSkill, that uses a differentiable physics simulator for skill abstraction to solve long-horizon deformable object manipulation tasks from sensory observations. In particular, we first obtain short-horizon skills using individual tools from a gradient-based optimizer, using the full state information in a differentiable simulator; we then learn a neural skill abstractor from the demonstration trajectories which takes RGBD images as input. Finally, we plan over the skills by finding the intermediate goals and then solve long-horizon tasks. We show the advantages of our method in a new set of sequential deformable object manipulation tasks compared to previous reinforcement learning algorithms and compared to the trajectory optimizer. Videos are available at our project page 1 . * This work was done during an internship at the MIT-IBM Watson AI Lab. | DIFFSKILL: SKILL ABSTRACTION FROM DIFFEREN- TIABLE PHYSICS FOR DEFORMABLE OBJECT MANIP- ULATIONS WITH TOOLS |
d257353677 | Intent detection with semantically similar fine-grained intents is a challenging task. To address it, we reformulate intent detection as a question-answering retrieval task by treating utterances and intent names as questions and answers. To that end, we utilize a question-answering retrieval architecture and adopt a two stages training schema with batch contrastive loss. In the pre-training stage, we improve query representations through self-supervised training. Then, in the finetuning stage, we increase contextualized token-level similarity scores between queries and answers from the same intent. Our results on three few-shot intent detection benchmarks achieve state-of-the-art performance. | Published as a conference paper at ICLR 2023 QAID: QUESTION ANSWERING INSPIRED FEW-SHOT INTENT DETECTION |
d207852415 | Recent trends of incorporating attention mechanisms in vision have led researchers to reconsider the supremacy of convolutional layers as a primary building block. Beyond helping CNNs to handle long-range dependencies,Ramachandran et al. (2019)showed that attention can completely replace convolution and achieve state-of-the-art performance on vision tasks. This raises the question: do learned attention layers operate similarly to convolutional layers? This work provides evidence that attention layers can perform convolution and, indeed, they often learn to do so in practice. Specifically, we prove that a multi-head self-attention layer with sufficient number of heads is at least as expressive as any convolutional layer. Our numerical experiments then show that self-attention layers attend to pixel-grid patterns similarly to CNN layers, corroborating our analysis. Our code is publicly available 1 .Published as a conference paper at ICLR 2020Contributions. In this work, we put forth theoretical and empirical evidence that self-attention layers can (and do) learn to behave similar to convolutional layers: I. From a theoretical perspective, we provide a constructive proof showing that self-attention layers can express any convolutional layers. Specifically, we show that a single multi-head self-attention layer using relative positional encoding can be re-parametrized to express any convolutional layer.II. Our experiments show that the first few layers of attention-only architectures (Ramachandran et al., 2019) do learn to attend on grid-like pattern around each query pixel, similar to our theoretical construction.Strikingly, this behavior is confirmed both for our quadratic encoding, but also for relative encoding that is learned. Our results seem to suggest that localized convolution is the right inductive bias for the first few layers of an image classifying network. We provide an interactive website 2 to explore how self-attention exploits localized position-based attention in lower layers and contentbased attention in deeper layers. For reproducibility purposes, our code is publicly available. | Published as a conference paper at ICLR 2020 ON THE RELATIONSHIP BETWEEN SELF-ATTENTION AND CONVOLUTIONAL LAYERS |
d5919904 | Current state-of-the-art classification and detection algorithms rely on supervised training. In this work we study unsupervised feature learning in the context of temporally coherent video data. We focus on feature learning from unlabeled video data, using the assumption that adjacent video frames contain semantically similar information. This assumption is exploited to train a convolutional pooling auto-encoder regularized by slowness and sparsity. We establish a connection between slow feature learning to metric learning and show that the trained encoder can be used to define a more temporally and semantically coherent metric. | UNSUPERVISED FEATURE LEARNING FROM TEMPO- RAL DATA |
d7071211 | Machine learning and deep learning in particular has advanced tremendously on perceptual tasks in recent years. However, it remains vulnerable against adversarial perturbations of the input that have been crafted specifically to fool the system while being quasi-imperceptible to a human. In this work, we propose to augment deep neural networks with a small "detector" subnetwork which is trained on the binary classification task of distinguishing genuine data from data containing adversarial perturbations. Our method is orthogonal to prior work on addressing adversarial perturbations, which has mostly focused on making the classification network itself more robust. We show empirically that adversarial perturbations can be detected surprisingly well even though they are quasi-imperceptible to humans. Moreover, while the detectors have been trained to detect only a specific adversary, they generalize to similar and weaker adversaries. In addition, we propose an adversarial attack that fools both the classifier and the detector and a novel training procedure for the detector that counteracts this attack. | Published as a conference paper at ICLR 2017 ON DETECTING ADVERSARIAL PERTURBATIONS |
d1896402 | Discourse relations bind smaller linguistic elements into coherent texts. However, automatically identifying discourse relations is difficult, because it requires understanding the semantics of the linked sentences. A more subtle challenge is that it is not enough to represent the meaning of each sentence of a discourse relation, because the relation may depend on links between lower-level elements, such as entity mentions. Our solution computes distributional meaning representations by composition up the syntactic parse tree. A key difference from previous work on compositional distributional semantics is that we also compute representations for entity mentions, using a novel downward compositional pass. Discourse relations are predicted not only from the distributional representations of the sentences, but also of their coreferent entity mentions. The resulting system obtains substantial improvements over the previous state-of-the-art in predicting implicit discourse relations in the Penn Discourse Treebank. | ENTITY-AUGMENTED DISTRIBUTIONAL SEMANTICS FOR DISCOURSE RELATIONS |
d252907242 | Recent text-to-image generation models have shown promising results in generating high-fidelity photo-realistic images. Though the results are astonishing to human eyes, how applicable these generated images are for recognition tasks remains under-explored. In this work, we extensively study whether and how synthetic images generated from state-of-the-art text-to-image generation models can be used for image recognition tasks, and focus on two perspectives: synthetic data for improving classification models in data-scarce settings (i.e. zero-shot and fewshot), and synthetic data for large-scale model pre-training for transfer learning. We showcase the powerfulness and shortcomings of synthetic data from existing generative models, and propose strategies for better applying synthetic data for recognition tasks. Code: https://github.com/CVMI-Lab/SyntheticData. * Part of the work is done during an internship at ByteDance. Email: ruifeihe@eee.hku.hk † Corresponding authors: songbai.site@gmail.com, xjqi@eee.hku.hk 1 At the beginning of this project, GLIDE is the only open-sourced text-to-image synthesis model that also delivers high-quality synthesis results.Published as a conference paper at ICLR 2023Our Findings. First, in the zero-shot setting, i.e. no real-world data are available, we demonstrate that synthetic data can significantly improve classification results on 17 diverse datasets: the performance is increased by 4.31% in top-1 accuracy on average, and even improved by as much as 17.86% on the EuroSAT dataset. To better leverage synthetic data in this setting, we also investigate useful strategies to increase data diversity, reduce data noise, and enhance data reliability. This is achieved by designing diversified text prompts and measuring the correlation of text and synthesized data with CLIP features.Second, in the few-shot setting, i.e. a few real images are available, albeit not as significant as in the zero-shot task, synthetic data are also shown to be beneficial and help us achieve a new state of the art. Our observation shows that the domain gap between synthetic data and downstream task data is one challenge on further improving the effectiveness of synthetic data on classifier learning. Fortunately, in this setting, the accessibility of real data samples can provide useful information about the data distribution of the downstream task. We thus propose to use real images as guidance in the generation process to reduce domain gaps and improve effectiveness.Third, in large-scale model pre-training for transfer learning, our study shows that synthetic data are suitable and effective for model pre-training, delivering superior transfer learning performance and even outperforming ImageNet pre-training. Especially, synthetic data work surprisingly well in unsupervised model pre-training, and favor ViT-based backbones. We also demonstrate that by increasing the label space (i.e. text prompts) for data generation, the enlarged data amount and diversity could further bring performance boosts. Besides, synthetic data can work collaboratively with real data (i.e. ImageNet) where we obtain improved performance when the model is initialized with ImageNet pre-trained weights.RELATED WORKSSynthetic Data for Image Recognition. There are mainly two forms of synthetic data for image recognition, i.e. 1) synthetic datasets generated from a traditional simulation pipeline; 2) synthetic images output from generative models.The first type, synthetic datasets(Dosovitskiy et al., 2015;Peng et al., 2017;Richter et al., 2016), are usually generated from a traditional pipeline with a specific data source, e.g.synthetic 2D renderings of 3D models or scenes from graphics engines. However, this traditional way of generating synthetic datasets has several drawbacks: 1) manually defined pipeline generated synthetic data may have a certain gap with real-world data; 2) taking up huge physical space to store and huge cost to share and transfer; 3) data amount and diversity bounded by the specific data source.Compared with synthetic datasets, generative models are a more efficient means of synthetic data representation, exhibiting favorable advantages: 1) could produce high-fidelity photorealistic images closer to real data since they are trained on real-world data; 2) highly condensed compared to synthetic data itself, and take up much reduced storage space; 3) potentially unlimited synthetic data size. Only recently, few works attempt to explore synthetic data generated from generative models for image recognition. Besnier et al. (2020) use a class-conditional GAN to train classifiers of the same classes. Zhang et al. (2021) leverage the latent code of StyleGAN (Karras et al., 2019) to produce labels for object part segmentation. While they achieve promising results, both works are task-wise and only employed on a small scale. Jahanian et al. (2021)use a GAN-based generator to generate multiple views to conduct unsupervised contrastive representation learning. These works, however, explore upon the traditional GAN-based models; in contrast, our work investigates with the best released text-to-image generation model, which demonstrates new customization ability for different downstream label space. | Published as a conference paper at ICLR 2023 IS SYNTHETIC DATA FROM GENERATIVE MODELS READY FOR IMAGE RECOGNITION? |
d246063533 | The ability to discover behaviours from past experience and transfer them to new tasks is a hallmark of intelligent agents acting sample-efficiently in the real world. Equipping embodied reinforcement learners with the same ability may be crucial for their successful deployment in robotics. While hierarchical and KL-regularized reinforcement learning individually hold promise here, arguably a hybrid approach could combine their respective benefits. Key to these fields is the use of information asymmetry across architectural modules to bias which skills are learnt. While asymmetry choice has a large influence on transferability, existing methods base their choice primarily on intuition in a domain-independent, potentially suboptimal, manner. In this paper, we theoretically and empirically show the crucial expressivity-transferability trade-off of skills across sequential tasks, controlled by information asymmetry. Given this insight, we introduce Attentive Priors for Expressive and Transferable Skills (APES), a hierarchical KL-regularized method, heavily benefiting from both priors and hierarchy. Unlike existing approaches, APES automates the choice of asymmetry by learning it in a data-driven, domaindependent, way based on our expressivity-transferability theorems. Experiments over complex transfer domains of varying levels of extrapolation and sparsity, such as robot block stacking, demonstrate the criticality of the correct asymmetric choice, with APES drastically outperforming previous methods.Published as a conference paper at ICLR 2023 independent to, and shared across, global coordinate frames. For manipulation, by not conditioning on certain object information, such as shape or weight, the robot learns generalisable grasping independent to these factors. Therefore, IA biases learnt behaviours and how they transfer across environments. Previous works predefined their IAs, which were primarily chosen on intuition and independent of domain. In addition, previously explored asymmetries were narrow(Table 1), which if sub-optimal, limit transfer benefits. We demonstrate that this indeed is the case for many methods on our domains(Galashov et al., 2019;Bagatella et al., 2022;Tirumala et al., 2019;Pertsch et al., 2021;Wulfmeier et al., 2019). A more systematic, theoretically and data driven, domain dependent, approach for choosing IA is thus required to maximally benefit from skills for transfer learning.In this paper, we employ hierarchical KL-regularized RL to effectively transfer skills across sequential tasks. We begin by theoretically and empirically showing the crucial expressivity-transferability trade-off, controlled by choice of IA, of skills across sequential tasks for hierarchical KL-regularized RL. Our expressivity-transferability theorems state that conditioning skill modules on too little or too much information, such as the current observation or entire history, can both be detrimental for transfer, due to the discovery of skills that are either too general (e.g. motor primitives) or too specialised (e.g. non-transferable task-level). We demonstrate this by ablating over a wide range of asymmetries between the hierarchical policy and prior. We show the inefficiencies of previous methods that choose highly sub-optimal IAs for our domains, drastically limiting transfer performance. Given this insight, we introduce APES, 'Attentive Priors for Expressive and Transferable Skills' as a method that forgoes user intuition and automates the choice of IA in a data driven, domain dependent, manner. APES builds on our expressivity-transferability theorems to learn the choice of asymmetry between policy and prior. Specifically, APES conditions the prior on the entire history, allowing for expressive skills to be discovered, and learns a low-entropic attention-mask over the input, paying attention only where necessary, to minimise covariate shift and improve transferability across domains. Experiments over domains of varying levels of sparsity and extrapolation, including a complex robot block stacking one, demonstrate APES' consistent superior performance over existing methods, whilst automating IA choice and by-passing arduous IA sweeps. Further ablations show the importance of combining hierarchy and priors for discovering expressive multi-modal behaviours. | Published as a conference paper at ICLR 2023 PRIORS, HIERARCHY, AND INFORMATION ASYMMETRY FOR SKILL TRANSFER IN REINFORCEMENT LEARNING |
d12200731 | A standard approach to Collaborative Filtering (CF), i.e. prediction of user ratings on items, relies on Matrix Factorization techniques. Representations for both users and items are computed from the observed ratings and used for prediction. Unfortunatly, these transductive approaches cannot handle the case of new users arriving in the system, with no known rating, a problem known as user cold-start. A common approach in this context is to ask these incoming users for a few initialization ratings. This paper presents a model to tackle this twofold problem of (i) finding good questions to ask, (ii) building efficient representations from this small amount of information. The model can also be used in a more standard (warm) context. Our approach is evaluated on the classical CF problem and on the cold-start problem on four different datasets showing its ability to improve baseline performance in both cases. | REPRESENTATION LEARNING FOR COLD-START RECOMMENDATION |
d231924480 | The emerging paradigm of federated learning (FL) strives to enable collaborative training of deep models on the network edge without centrally aggregating raw data and hence improving data privacy. In most cases, the assumption of independent and identically distributed samples across local clients does not hold for federated learning setups. Under this setting, neural network training performance may vary significantly according to the data distribution and even hurt training convergence. Most of the previous work has focused on a difference in the distribution of labels or client shifts. Unlike those settings, we address an important problem of FL, e.g., different scanners/sensors in medical imaging, different scenery distribution in autonomous driving (highway vs. city), where local clients store examples with different distributions compared to other clients, which we denote as feature shift non-iid. In this work, we propose an effective method that uses local batch normalization to alleviate the feature shift before averaging models. The resulting scheme, called FedBN, outperforms both classical FedAvg, as well as the state-of-the-art for non-iid data (FedProx) on our extensive experiments. These empirical results are supported by a convergence analysis that shows in a simplified setting that FedBN has a faster convergence rate than FedAvg. Code is available at https://github.com/med-air/FedBN.Recent studies have attempted to address the problem of FL on non-iid data. Most variants of FedAvg primarily tackle the issues of stability, client drift and heterogeneous label distribution over | Published as a conference paper at ICLR 2021 FEDBN: FEDERATED LEARNING ON NON-IID FEATURES VIA LOCAL BATCH NORMALIZATION |
d211532533 | Fine-tuning from pre-trained ImageNet models has become the de-facto standard for various computer vision tasks. Current practices for fine-tuning typically involve selecting an ad-hoc choice of hyperparameters and keeping them fixed to values normally used for training from scratch. This paper re-examines several common practices of setting hyperparameters for fine-tuning. Our findings are based on extensive empirical evaluation for fine-tuning on various transfer learning benchmarks. (1) While prior works have thoroughly investigated learning rate and batch size, momentum for fine-tuning is a relatively unexplored parameter. We find that the value of momentum also affects fine-tuning performance and connect it with previous theoretical findings. (2) Optimal hyperparameters for fine-tuning, in particular, the effective learning rate, are not only dataset dependent but also sensitive to the similarity between the source domain and target domain. This is in contrast to hyperparameters for training from scratch. (3) Reference-based regularization that keeps models close to the initial model does not necessarily apply for "dissimilar" datasets. Our findings challenge common practices of finetuning and encourages deep learning practitioners to rethink the hyperparameters for fine-tuning. | Published as a conference paper at ICLR 2020 RETHINKING THE HYPERPARAMETERS FOR FINE-TUNING |
d232240418 | Recent research in dynamic convolution shows substantial performance boost for efficient CNNs, due to the adaptive aggregation of K static convolution kernels. It has two limitations: (a) it increases the number of convolutional weights by Ktimes, and (b) the joint optimization of dynamic attention and static convolution kernels is challenging. In this paper, we revisit it from a new perspective of matrix decomposition and reveal the key issue is that dynamic convolution applies dynamic attention over channel groups after projecting into a higher dimensional latent space. To address this issue, we propose dynamic channel fusion to replace dynamic attention over channel groups. Dynamic channel fusion not only enables significant dimension reduction of the latent space, but also mitigates the joint optimization difficulty. As a result, our method is easier to train and requires significantly fewer parameters without sacrificing accuracy. Source code is at https://github.com/liyunsheng13/dcd. | Published as a conference paper at ICLR 2021 REVISITING DYNAMIC CONVOLUTION VIA MATRIX DECOMPOSITION |
d6981893 | The ability to plan and execute goal specific actions in varied and unseen environments settings is a central requirement of intelligent agents. In this paper, we explore how an agent can be equipped with an internal model of the dynamics of the external world, and how it can use this model to plan novel actions by running multiple internal simulations ("visual imagination"). Our models directly process raw visual input, and use a novel object-centric prediction formulation based on visual glimpses centered on objects (fixations) to enforce translational invariance of the learned physical laws. The agent trains itself through random interaction with a collection of different environments, and the resulting model can then be used to plan goal-directed actions in novel environments that were previously never encountered by the agent. We demonstrate that our agent can accurately plan actions for playing a simulated billiards game, which requires pushing a ball into a target position or into collision with another ball. * equal contribution 1 arXiv:1511.07404v3 [cs.CV] | LEARNING VISUAL PREDICTIVE MODELS OF PHYSICS FOR PLAYING BILLIARDS |
d247083956 | How can we measure the reasoning capabilities of intelligence systems? Visual question answering provides a convenient framework for testing the model's abilities by interrogating the model through questions about the scene. However, despite scores of various visual QA datasets and architectures, which sometimes yield even a super-human performance, the question of whether those architectures can actually reason remains open to debate. To answer this, we extend the visual question answering framework and propose the following behavioral test in the form of a two-player game. We consider black-box neural models of CLEVR. These models are trained on a diagnostic dataset benchmarking reasoning. Next, we train an adversarial player that re-configures the scene to fool the CLEVR model. We show that CLEVR models, which otherwise could perform at a "human level", can easily be fooled by our agent. Our results put in doubt whether data-driven approaches can do reasoning without exploiting the numerous biases that are often present in those datasets. Finally, we also propose a controlled experiment measuring the efficiency of such models to learn and perform reasoning. . Learning visual question answering by bootstrapping hard attention. In . Film: Visual reasoning with a general conditioning layer. In AAAI, 2018.Oskar Pfungst. Clever Hans:(the horse of Mr. Von Osten.) a contribution to experimental animal and human psychology. Holt, Rinehart and Winston, 1911.Piotr Piekos, Henryk Michalewski, and Mateusz Malinowski. Measuring and improving bert's mathematical abilities by predicting the order of reasoning. | Published as a conference paper at ICLR 2022 MEASURING CLEVRNESS: BLACK-BOX TESTING OF VISUAL REASONING MODELS |
d252780503 | Warning: this paper contains model outputs exhibiting offensiveness and biases.Recently pre-trained language models (PLMs) have prospered in various natural language generation (NLG) tasks due to their ability to generate fairly fluent text. Nevertheless, these models are observed to capture and reproduce harmful contents in training corpora, typically toxic language and social biases, raising severe moral issues. Prior works on ethical NLG tackle detoxifying and debiasing separately, which is problematic since we find debiased models still exhibit toxicity while detoxified ones even exacerbate social biases. To address such a challenge, we propose the first unified framework of detoxifying and debiasing called UD-DIA, which jointly formalizes these two problems as rectifying the output space. We theoretically interpret our framework as learning a text distribution mixing weighted attributes. Besides, UDDIA conducts adaptive optimization of only a few parameters during decoding based on a parameter-efficient tuning schema without any training data. This leads to minimal generation quality loss and improved rectification performance with acceptable computational cost. Experimental results demonstrate that compared to several strong baselines, UDDIA achieves debiasing and detoxifying simultaneously and better balances efficiency and effectiveness, taking a further step towards practical ethical NLG. * Work done during an internship at MSRA. yangzh20@mails.tsinghua.edu.cn † Corresponding authors: P. Li | Published as a conference paper at ICLR 2023 UNIFIED DETOXIFYING AND DEBIASING IN LAN- GUAGE GENERATION VIA INFERENCE-TIME ADAPTIVE OPTIMIZATION |
d231812484 | Interpretation of Deep Neural Networks (DNNs) training as an optimal control problem with nonlinear dynamical systems has received considerable attention recently, yet the algorithmic development remains relatively limited. In this work, we make an attempt along this line by reformulating the training procedure from the trajectory optimization perspective. We first show that most widely-used algorithms for training DNNs can be linked to the Differential Dynamic Programming (DDP), a celebrated second-order method rooted in the Approximate Dynamic Programming. In this vein, we propose a new class of optimizer, DDP Neural Optimizer (DDP-NOpt), for training feedforward and convolution networks. DDPNOpt features layer-wise feedback policies which improve convergence and reduce sensitivity to hyper-parameter over existing methods. It outperforms other optimal-control inspired training methods in both convergence and complexity, and is competitive against state-of-the-art first and second order methods. We also observe DDPNOpt has surprising benefit in preventing gradient vanishing. Our work opens up new avenues for principled algorithmic design built upon the optimal control theory. | Published as a conference paper at ICLR 2021 DDPNOPT: DIFFERENTIAL DYNAMIC PROGRAMMING NEURAL OPTIMIZER |
d257255438 | Pre-training representations (a.k.a. foundation models) has recently become a prevalent learning paradigm, where one first pre-trains a representation using large-scale unlabeled data, and then learns simple predictors on top of the representation using small labeled data from the downstream tasks. There are two key desiderata for the representation: label efficiency (the ability to learn an accurate classifier on top of the representation with a small amount of labeled data) and universality (usefulness across a wide range of downstream tasks). In this paper, we focus on one of the most popular instantiations of this paradigm: contrastive learning with linear probing, i.e., learning a linear predictor on the representation pre-trained by contrastive learning. We show that there exists a trade-off between the two desiderata so that one may not be able to achieve both simultaneously. Specifically, we provide analysis using a theoretical data model and show that, while more diverse pre-training data result in more diverse features for different tasks (improving universality), it puts less emphasis on task-specific features, giving rise to larger sample complexity for down-stream supervised tasks, and thus worse prediction performance. Guided by this analysis, we propose a contrastive regularization method to improve the trade-off. We validate our analysis and method empirically with systematic experiments using real-world datasets and foundation models. | THE TRADE-OFF BETWEEN UNIVERSALITY AND LA- BEL EFFICIENCY OF REPRESENTATIONS FROM CON- TRASTIVE LEARNING |
d56657907 | We investigate a variant of variational autoencoders where there is a superstructure of discrete latent variables on top of the latent features. In general, our superstructure is a tree structure of multiple super latent variables and it is automatically learned from data. When there is only one latent variable in the superstructure, our model reduces to one that assumes the latent features to be generated from a Gaussian mixture model. We call our model the latent tree variational autoencoder (LTVAE). Whereas previous deep learning methods for clustering produce only one partition of data, LTVAE produces multiple partitions of data, each being given by one super latent variable. This is desirable because high dimensional data usually have many different natural facets and can be meaningfully partitioned in multiple ways. | LEARNING LATENT SUPERSTRUCTURES IN VARIA- TIONAL AUTOENCODERS FOR DEEP MULTIDIMEN- SIONAL CLUSTERING |
d235614315 | We propose a novel federated learning method for distributively training neural network models, where the server orchestrates cooperation between a subset of randomly chosen devices in each round. We view Federated Learning problem primarily from a communication perspective and allow more device level computations to save transmission costs. We point out a fundamental dilemma, in that the minima of the local-device level empirical loss are inconsistent with those of the global empirical loss. Different from recent prior works, that either attempt inexact minimization or utilize devices for parallelizing gradient computation, we propose a dynamic regularizer for each device at each round, so that in the limit the global and device solutions are aligned. We demonstrate both through empirical results on real and synthetic data as well as analytical results that our scheme leads to efficient training, in both convex and non-convex settings, while being fully agnostic to device heterogeneity and robust to large number of devices, partial participation and unbalanced data. | Published as a conference paper at ICLR 2021 FEDERATED LEARNING BASED ON DYNAMIC REGULARIZATION |
d3649599 | We present a Neural Program Search, an algorithm to generate programs from natural language description and a small number of input / output examples. The algorithm combines methods from Deep Learning and Program Synthesis fields by designing rich domain-specific language (DSL) and defining efficient search algorithm guided by a Seq2Tree model on it. To evaluate the quality of the approach we also present a semi-synthetic dataset of descriptions with test examples and corresponding programs. We show that our algorithm significantly outperforms a sequence-to-sequence model with attention baseline. | Workshop track -ICLR 2018 Neural Program Search: Solving Programming Tasks from Description and Examples |
d246275797 | Combinatorial optimization lies at the core of many real-world problems. Especially since the rise of graph neural networks (GNNs), the deep learning community has been developing solvers that derive solutions to NP-hard problems by learning the problem-specific solution structure. However, reproducing the results of these publications proves to be difficult. We make three contributions. First, we present an open-source benchmark suite for the NP-hard MAXIMUM INDEPENDENT SET problem, in both its weighted and unweighted variants. The suite offers a unified interface to various state-of-the-art traditional and machine learning-based solvers. Second, using our benchmark suite, we conduct an in-depth analysis of the popular guided tree search algorithm by Li et al. [NeurIPS 2018], testing various configurations on small and large synthetic and real-world graphs. By re-implementing their algorithm with a focus on code quality and extensibility, we show that the graph convolution network used in the tree search does not learn a meaningful representation of the solution structure, and can in fact be replaced by random values. Instead, the tree search relies on algorithmic techniques like graph kernelization to find good solutions. Thus, the results from the original publication are not reproducible. Third, we extend the analysis to compare the tree search implementations to other solvers, showing that the classical algorithmic solvers often are faster, while providing solutions of similar quality. Additionally, we analyze a recent solver based on reinforcement learning and observe that for this solver, the GNN is responsible for the competitive solution quality.Published as a conference paper at ICLR 2022 reason for the interest in initiatives such as Papers With Code 1 and the ReScience Journal 2 . Keeping the importance of reproducibility in mind, we evaluate various machine learning approaches for combinatorial optimization, and compare them to traditional solvers. We focus on the MAXIMUM (WEIGHTED) INDEPENDENT SET problem (Miller & Muller, 1960; Karp, 1972), as the previous works we analyze have centered around variants of this problem. Overall, we contribute the following. • We provide an open-source, extensible benchmark suite for MAXIMUM (WEIGHTED) INDEPENDENT SET solvers. The software currently supports five state-of-the-art solvers, INTEL-TREESEARCH (Li, Chen, and Koltun, 2018), GUROBI (Gurobi Optimization LLC, 2021), KAMIS (Lamm et al., 2017; Hespe et al., 2019; Lamm et al., 2019), LEARNING WHAT TO DEFER (Ahn, Seo, and Shin, 2020), and DGL-TREESEARCH. Our DGL-TREESEARCH is a modern re-implementation of the INTEL-TREESEARCH, implemented in PyTorch (Paszke et al., 2019) and the Deep Graph Library (Wang et al., 2019), with a focus on clean, readable code, as well as performance, and it fixes various issues of the original code. Our evaluation suite lays the ground for further research on hard combinatorial (graph) problems and aims at providing a fair and comparable environment for further evaluations. • Using our re-implementation of the tree search, we propose and analyze additional techniques aiming at improving the guided search. Employing the benchmark suite, we conduct an exhaustive analysis of various configurations of the tree search algorithms, showing that the results of the highly-cited INTEL-TREESEARCH approach are not reproducible, neither with the original code nor with our re-implementation. When exploring the design space further, we show that the various techniques used by the tree search algorithm to improve the results, like graph kernelization, are the reason for good performance, especially on hard data sets. In fact, replacing the GNN output with random values performs similar to using the trained network. • Having analyzed the configuration space, we compare the tree search approaches to the classical solvers like GUROBI and KAMIS, showing that problem-tailored solvers are often the superior approach. Without using techniques like graph reduction in the tree search, classical solvers are superior. The classical solvers show to be more efficient even when accessing these routines -that have been implemented for the algorithmic solvers in the first place -in the tree search. Last, we show that LEARNING WHAT TO DEFER seems to be able to find good results very quickly, indicating that unsupervised reinforcement learning for combinatorial problems is a promising direction for future research.INDEPENDENT SET SOLVERS AND MACHINE LEARNING FOR COMBINATORIAL OPTIMIZATIONIn this section, we formally introduce the MAXIMUM INDEPENDENT SET (MIS) problem as well as the solvers included in our analysis, and discuss related work in the broader space of deep learning for combinatorial optimization. Given an undirected graph G = (V, E), an independent set is a set of vertices S ⊆ V for which for all vertices u, v ∈ S, (u, v) ∈ E. For u ∈ V , let w u be its weight, and let IS(G) be all independent sets of G, then the MAXIMUM WEIGHTED INDEPENDENT SET (MWIS) problem aims at determining arg max S∈IS(G) u∈S w u . The unweighted MIS problem is equivalent to the MWIS problem, where f.a. u ∈ V , w u = 1. Both problems are strongly NP-complete (Garey & Johnson, 1978). Next, we briefly explain the solvers that we use to find such MIS.Gurobi. GUROBI is a commercial mathematical optimization solver. There are various ways of formulating the M(W)IS problem mathematically(Butenko, 2003). In the main paper, we formulate the MWIS problem as the linear program above, and discuss other variants in Appendix D.KaMIS.KAMIS is an open-source solver tailored towards the MIS and MWIS problems. It offers support both for the unweighted case (Lamm et al., 2017; Hespe et al., 2019) as well as the weighted case (Lamm et al., 2019). It employs graph kernelization and an optimized branch-and-bound algorithm to efficiently find independent sets. Note that the algorithms and techniques differ between the weighted and unweighted cases. We use the code unmodified from the official repository 3 . | Published as a conference paper at ICLR 2022 WHAT'S WRONG WITH DEEP LEARNING IN TREE SEARCH FOR COMBINATORIAL OPTIMIZATION |
d3649804 | This paper presents a framework to tackle combinatorial optimization problems using neural networks and reinforcement learning. We focus on the traveling salesman problem (TSP) and train a recurrent neural network that, given a set of city coordinates, predicts a distribution over different city permutations. Using negative tour length as the reward signal, we optimize the parameters of the recurrent neural network using a policy gradient method. We compare learning the network parameters on a set of training graphs against learning them on individual test graphs. Despite the computational expense, without much engineering and heuristic designing, Neural Combinatorial Optimization achieves close to optimal results on 2D Euclidean graphs with up to 100 nodes. Applied to the KnapSack, another NP-hard problem, the same method obtains optimal solutions for instances with up to 200 items. * Equal contributions. Members of the Google Brain Residency program (g.co/brainresidency). | NEURAL COMBINATORIAL OPTIMIZATION WITH REINFORCEMENT LEARNING |
d212859361 | Over-fitting and over-smoothing are two main obstacles of developing deep Graph Convolutional Networks (GCNs) for node classification. In particular, over-fitting weakens the generalization ability on small dataset, while over-smoothing impedes model training by isolating output representations from the input features with the increase in network depth. This paper proposes DropEdge, a novel and flexible technique to alleviate both issues. At its core, DropEdge randomly removes a certain number of edges from the input graph at each training epoch, acting like a data augmenter and also a message passing reducer. Furthermore, we theoretically demonstrate that DropEdge either reduces the convergence speed of over-smoothing or relieves the information loss caused by it. More importantly, our DropEdge is a general skill that can be equipped with many other backbone models (e.g. GCN, ResGCN, GraphSAGE, and JKNet) for enhanced performance. Extensive experiments on several benchmarks verify that DropEdge consistently improves the performance on a variety of both shallow and deep GCNs. The effect of DropEdge on preventing over-smoothing is empirically visualized and validated as well. Codes are released on https://github.com/DropEdge/DropEdge. | DROPEDGE: TOWARDS DEEP GRAPH CONVOLU- TIONAL NETWORKS ON NODE CLASSIFICATION |
d3514022 | Model-free reinforcement learning (RL) is a powerful, general tool for learning complex behaviors. However, its sample efficiency is often impractically large for solving challenging real-world problems, even with off-policy algorithms such as Q-learning. A limiting factor in classic model-free RL is that the learning signal consists only of scalar rewards, ignoring much of the rich information contained in state transition tuples. Model-based RL uses this information, by training a predictive model, but often does not achieve the same asymptotic performance as model-free RL due to model bias. We introduce temporal difference models (TDMs), a family of goal-conditioned value functions that can be trained with model-free learning and used for model-based control. TDMs combine the benefits of model-free and model-based RL: they leverage the rich information in state transitions to learn very efficiently, while still attaining asymptotic performance that exceeds that of direct model-based RL methods. Our experimental results show that, on a range of continuous control tasks, TDMs provide a substantial improvement in efficiency compared to state-of-the-art model-based and model-free methods. * denotes equal contribution | Published as a conference paper at ICLR 2018 TEMPORAL DIFFERENCE MODELS: MODEL-FREE DEEP RL FOR MODEL-BASED CONTROL |
d252683963 | Despite the widespread use of unsupervised models, very few methods are designed to explain them. Most explanation methods explain a scalar model output. However, unsupervised models output representation vectors, the elements of which are not good candidates to explain because they lack semantic meaning. To bridge this gap, recent works defined a scalar explanation output: a dot product-based similarity in the representation space to the sample being explained (i.e., an explicand). Although this enabled explanations of unsupervised models, the interpretation of this approach can still be opaque because similarity to the explicand's representation may not be meaningful to humans. To address this, we propose contrastive corpus similarity, a novel and semantically meaningful scalar explanation output based on a reference corpus and a contrasting foil set of samples. We demonstrate that contrastive corpus similarity is compatible with many post-hoc feature attribution methods to generate COntrastive COrpus Attributions (COCOA) and quantitatively verify that features important to the corpus are identified. We showcase the utility of COCOA in two ways: (i) we draw insights by explaining augmentations of the same image in a contrastive learning setting (SimCLR); and (ii) we perform zero-shot object localization by explaining the similarity of image representations to jointly learned text representations (CLIP). * Equal contribution. | Published as a conference paper at ICLR 2023 CONTRASTIVE CORPUS ATTRIBUTION FOR EXPLAIN- ING REPRESENTATIONS |
d252668812 | Many Graph Neural Networks (GNNs) perform poorly compared to simple heuristics on Link Prediction (LP) tasks. This is due to limitations in expressive power such as the inability to count triangles (the backbone of most LP heuristics) and because they can not distinguish automorphic nodes (those having identical structural roles). Both expressiveness issues can be alleviated by learning link (rather than node) representations and incorporating structural features such as triangle counts. Since explicit link representations are often prohibitively expensive, recent works resorted to subgraph-based methods, which have achieved state-ofthe-art performance for LP, but suffer from poor efficiency due to high levels of redundancy between subgraphs. We analyze the components of subgraph GNN (SGNN) methods for link prediction. Based on our analysis, we propose a novel full-graph GNN called ELPH (Efficient Link Prediction with Hashing) that passes subgraph sketches as messages to approximate the key components of SGNNs without explicit subgraph construction. ELPH is provably more expressive than Message Passing GNNs (MPNNs). It outperforms existing SGNN models on many standard LP benchmarks while being orders of magnitude faster. However, it shares the common GNN limitation that it is only efficient when the dataset fits in GPU memory. Accordingly, we develop a highly scalable model, called BUDDY, which uses feature precomputation to circumvent this limitation without sacrificing predictive performance. Our experiments show that BUDDY also outperforms SGNNs on standard LP benchmarks while being highly scalable and faster than ELPH. * Equal contribution. † Work done while at Twitter Inc. ‡ benjamin.chamberlain@gmail.com. | Graph Neural Networks for Link Prediction with Subgraph Sketching Charm Therapeutics |
d211126925 | While deep neural networks have achieved impressive performance on a range of NLP tasks, these data-hungry models heavily rely on labeled data, which restricts their applications in scenarios where data annotation is expensive. Natural language (NL) explanations have been demonstrated very useful additional supervision, which can provide sufficient domain knowledge for generating more labeled data over new instances, while the annotation time only doubles. However, directly applying them for augmenting model learning encounters two challenges: (1) NL explanations are unstructured and inherently compositional, which asks for a modularized model to represent their semantics, (2) NL explanations often have large numbers of linguistic variants, resulting in low recall and limited generalization ability. In this paper, we propose a novel Neural Execution Tree (NExT) framework 1 to augment training data for text classification using NL explanations. After transforming NL explanations into executable logical forms by semantic parsing, NExT generalizes different types of actions specified by the logical forms for labeling data instances, which substantially increases the coverage of each NL explanation. Experiments on two NLP tasks (relation extraction and sentiment analysis) demonstrate its superiority over baseline methods. Its extension to multi-hop question answering achieves performance gain with light annotation effort. * Equal contribution. The order is decided by a coin toss. The work was done when visiting USC. | Published as a conference paper at ICLR 2020 LEARNING FROM EXPLANATIONS WITH NEURAL EXECUTION TREE |
d257496463 | Non-autoregressive translation (NAT) reduces the decoding latency but suffers from performance degradation due to the multi-modality problem. Recently, the structure of directed acyclic graph has achieved great success in NAT, which tackles the multi-modality problem by introducing dependency between vertices. However, training it with negative log-likelihood loss implicitly requires a strict alignment between reference tokens and vertices, weakening its ability to handle multiple translation modalities. In this paper, we hold the view that all paths in the graph are fuzzily aligned with the reference sentence. We do not require the exact alignment but train the model to maximize a fuzzy alignment score between the graph and reference, which takes captured translations in all modalities into account. Extensive experiments on major WMT benchmarks show that our method substantially improves translation performance and increases prediction confidence, setting a new state of the art for NAT on the raw training data. 1 | Published as a conference paper at ICLR 2023 FUZZY ALIGNMENTS IN DIRECTED ACYCLIC GRAPH FOR NON-AUTOREGRESSIVE MACHINE TRANSLATION |
d6856808 | Deep neural networks can be obscenely wasteful. When processing video, a convolutional network expends a fixed amount of computation for each frame with no regard to the similarity between neighbouring frames. As a result, it ends up repeatedly doing very similar computations. To put an end to such waste, we introduce Sigma-Delta networks. With each new input, each layer in this network sends a discretized form of its change in activation to the next layer. Thus the amount of computation that the network does scales with the amount of change in the input and layer activations, rather than the size of the network. We introduce an optimization method for converting any pre-trained deep network into an optimally efficient Sigma-Delta network, and show that our algorithm, if run on the appropriate hardware, could cut at least an order of magnitude from the computational cost of processing video data. | SIGMA-DELTA QUANTIZED NETWORKS |
d18233038 | Convolutional networks are one of the most widely employed architectures in computer vision and machine learning. In order to leverage their ability to learn complex functions, large amounts of data are required for training. Training a large convolutional network to produce state-of-the-art results can take weeks, even when using modern GPUs. Producing labels using a trained network can also be costly when dealing with web-scale datasets. In this work, we present a simple algorithm which accelerates training and inference by a significant factor, and can yield improvements of over an order of magnitude compared to existing state-of-the-art implementations. This is done by computing convolutions as pointwise products in the Fourier domain while reusing the same transformed feature map many times. The algorithm is implemented on a GPU architecture and addresses a number of related challenges. | Fast Training of Convolutional Networks through FFTs |
d259298732 | Learning policies from previously recorded data is a promising direction for realworld robotics tasks, as online learning is often infeasible. Dexterous manipulation in particular remains an open problem in its general form. The combination of offline reinforcement learning with large diverse datasets, however, has the potential to lead to a breakthrough in this challenging domain analogously to the rapid progress made in supervised learning in recent years. To coordinate the efforts of the research community toward tackling this problem, we propose a benchmark including: i) a large collection of data for offline learning from a dexterous manipulation platform on two tasks, obtained with capable RL agents trained in simulation; ii) the option to execute learned policies on a real-world robotic system and a simulation for efficient debugging. We evaluate prominent open-sourced offline reinforcement learning algorithms on the datasets and provide a reproducible experimental setup for offline reinforcement learning on real systems. Visit https://sites.google.com/view/ benchmarking-offline-rl-real for more details. . Transferring dexterous manipulation from GPU simulation to a remote real-world trifinger. , et al. Dota 2 with large scale deep reinforcement learning. arXiv preprint arXiv:1912.06680, 2019. Fabio Bonsignorio and Angel P. del Pobil. Toward replicable and measurable robotics research [from the guest editors]. : Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018. | Published as a conference paper at ICLR 2023 BENCHMARKING OFFLINE REINFORCEMENT LEARNING ON REAL-ROBOT HARDWARE |
d3476061 | For fast and energy-efficient deployment of trained deep neural networks on resource-constrained embedded hardware, each learned weight parameter should ideally be represented and stored using a single bit. Error-rates usually increase when this requirement is imposed. Here, we report large improvements in error rates on multiple datasets, for deep convolutional neural networks deployed with 1-bit-per-weight.Using wide residual networks as our main baseline, our approach simplifies existing methods that binarize weights by applying the sign function in training; we apply scaling factors for each layer with constant unlearned values equal to the layer-specific standard deviations used for initialization. For CIFAR-10, CIFAR-100 and ImageNet, and models with 1-bitper-weight requiring less than 10 MB of parameter memory, we achieve error rates of 3.9%, 18.5% and 26.0% / 8.5% (Top-1 / Top-5) respectively. We also considered MNIST, SVHN and ImageNet32, achieving 1-bit-per-weight test results of 0.27%, 1.9%, and 41.3% / 19.1% respectively. For CIFAR, our error rates halve previously reported values, and are within about 1% of our errorrates for the same network with full-precision weights. For networks that overfit, we also show significant improvements in error rate by not learning batch normalization scale and offset parameters. This applies to both full precision and 1-bit-per-weight networks. Using a warm-restart learning-rate schedule, we found that training for 1-bit-per-weight is just as fast as full-precision networks, with better accuracy than standard schedules, and achieved about 98%-99% of peak performance in just 62 training epochs for CIFAR-10/100. For full training code and trained models in MATLAB, Keras and PyTorch see https://github.com/McDonnell-Lab/1-bit-per-weight/.Published as a conference paper at ICLR 2018 ideal case of single-bit parameters and/or processing, still fall well-short of state-of-the-art error rates on important benchmarks. | Published as a conference paper at ICLR 2018 TRAINING WIDE RESIDUAL NETWORKS FOR DEPLOY- MENT USING A SINGLE BIT FOR EACH WEIGHT |
d251280102 | Normalizing flows are tractable density models that can approximate complicated target distributions, e.g. Boltzmann distributions of physical systems. However, current methods for training flows either suffer from mode-seeking behavior, use samples from the target generated beforehand by expensive MCMC methods, or use stochastic losses that have high variance. To avoid these problems, we augment flows with annealed importance sampling (AIS) and minimize the masscovering α-divergence with α = 2, which minimizes importance weight variance. Our method, Flow AIS Bootstrap (FAB), uses AIS to generate samples in regions where the flow is a poor approximation of the target, facilitating the discovery of new modes. We apply FAB to multimodal targets and show that we can approximate them very accurately where previous methods fail. To the best of our knowledge, we are the first to learn the Boltzmann distribution of the alanine dipeptide molecule using only the unnormalized target density, without access to samples generated via Molecular Dynamics (MD) simulations: FAB produces better results than training via maximum likelihood on MD samples while using 100 times fewer target evaluations. After reweighting the samples, we obtain unbiased histograms of dihedral angles that are almost identical to the ground truth. . Machine learning of coarse-grained molecular dynamics force fields. ACS central science, 5(5):755-767, 2019. Ronald J Williams. Simple statistical gradient-following algorithms for connectionist reinforcement learning. Machine learning, 8(3):229-256, 1992. | Published as a conference paper at ICLR 2023 FLOW ANNEALED IMPORTANCE SAMPLING BOOTSTRAP |
d3714278 | Several deep learning models have been proposed for question answering. However, due to their single-pass nature, they have no way to recover from local maxima corresponding to incorrect answers. To address this problem, we introduce the Dynamic Coattention Network (DCN) for question answering. The DCN first fuses co-dependent representations of the question and the document in order to focus on relevant parts of both. Then a dynamic pointing decoder iterates over potential answer spans. This iterative procedure enables the model to recover from initial local maxima corresponding to incorrect answers. On the Stanford question answering dataset, a single DCN model improves the previous state of the art from 71.0% F1 to 75.9%, while a DCN ensemble obtains 80.4% F1. | Published as a conference paper at ICLR 2017 DYNAMIC COATTENTION NETWORKS FOR QUESTION ANSWERING |
d17272965 | Despite the widespread practical success of deep learning methods, our theoretical understanding of the dynamics of learning in deep neural networks remains quite sparse. We attempt to bridge the gap between the theory and practice of deep learning by systematically analyzing learning dynamics for the restricted case of deep linear neural networks. Despite the linearity of their input-output map, such networks have nonlinear gradient descent dynamics on weights that change with the addition of each new hidden layer. We show that deep linear networks exhibit nonlinear learning phenomena similar to those seen in simulations of nonlinear networks, including long plateaus followed by rapid transitions to lower error solutions, and faster convergence from greedy unsupervised pretraining initial conditions than from random initial conditions. We provide an analytical description of these phenomena by finding new exact solutions to the nonlinear dynamics of deep learning. Our theoretical analysis also reveals the surprising finding that as the depth of a network approaches infinity, learning speed can nevertheless remain finite: for a special class of initial conditions on the weights, very deep networks incur only a finite, depth independent, delay in learning speed relative to shallow networks. We show that, under certain conditions on the training data, unsupervised pretraining can find this special class of initial conditions, while scaled random Gaussian initializations cannot. We further exhibit a new class of random orthogonal initial conditions on weights that, like unsupervised pre-training, enjoys depth independent learning times. We further show that these initial conditions also lead to faithful propagation of gradients even in deep nonlinear networks, as long as they operate in a special regime known as the edge of chaos.Deep learning methods have realized impressive performance in a range of applications, from visual object classification [1, 2, 3] to speech recognition [4] and natural language processing[5,6]. These successes have been achieved despite the noted difficulty of training such deep architectures[7,8,9,10,11]. Indeed, many explanations for the difficulty of deep learning have been advanced in the literature, including the presence of many local minima, low curvature regions due to saturating nonlinearities, and exponential growth or decay of back-propagated gradients[12,13,14,15]. Furthermore, many neural network simulations have observed 1 arXiv:1312.6120v3 [cs.NE] | Exact solutions to the nonlinear dynamics of learning in deep linear neural networks |
d252595924 | We develop ShiftMatch 1 , a new training-data-dependent likelihood for robustness to corruption in Bayesian neural networks (BNNs). ShiftMatch is inspired by the training-data-dependent "EmpCov" priors fromIzmailov et al. (2021a), and efficiently matches test-time spatial correlations to those at training time. Critically, ShiftMatch is designed to leave the neural network's training time likelihood unchanged, allowing it to use publicly available samples from pre-trained BNNs. Using pre-trained HMC samples, ShiftMatch gives strong performance improvements on CIFAR-10-C, outperforms EmpCov priors (though ShiftMatch uses extra information from a minibatch of corrupted test points), and is perhaps the first Bayesian method capable of convincingly outperforming plain deep ensembles. | Published as a conference paper at ICLR 2023 ROBUSTNESS TO CORRUPTION IN PRE-TRAINED BAYESIAN NEURAL NETWORKS |
d246705937 | Non-stationarity is one thorny issue in cooperative multi-agent reinforcement learning (MARL). One of the reasons is the policy changes of agents during the learning process. Some existing works have discussed various consequences caused by non-stationarity with several kinds of measurement indicators. This makes the objectives or goals of existing algorithms are inevitably inconsistent and disparate. In this paper, we introduce a novel notion, the δ-stationarity measurement, to explicitly measure the non-stationarity of a policy sequence, which can be further proved to be bounded by the KL-divergence of consecutive joint policies. A straightforward but highly non-trivial way is to control the joint policies' divergence, which is difficult to estimate accurately by imposing the trust-region constraint on the joint policy. Although it has lower computational complexity to decompose the joint policy and impose trust-region constraints on the factorized policies, simple policy factorization like mean-field approximation will lead to more considerable policy divergence, which can be considered as the trust-region decomposition dilemma. We model the joint policy as a pairwise Markov random field and propose a trust-region decomposition network (TRD-Net) based on message passing to estimate the joint policy divergence more accurately. The Multi-Agent Mirror descent policy algorithm with Trust region decomposition, called MAMT, is established by adjusting the trust-region of the local policies adaptively in an end-to-end manner. MAMT can approximately constrain the consecutive joint policies' divergence to satisfy δ-stationarity and alleviate the non-stationarity problem. Our method can bring noticeable and stable performance improvement compared with baselines in cooperative tasks of different complexity. | Published as a conference paper at ICLR 2022 DEALING WITH NON-STATIONARITY IN MARL VIA TRUST-REGION DECOMPOSITION |
d209500529 | Adaptive gradient algorithms perform gradient-based updates using the history of gradients and are ubiquitous in training deep neural networks. While adaptive gradient methods theory is well understood for minimization problems, the underlying factors driving their empirical success in min-max problems such as GANs remain unclear. In this paper, we aim at bridging this gap from both theoretical and empirical perspectives. First, we analyze a variant of Optimistic Stochastic Gradient (OSG) proposed in(Daskalakis et al., 2017)for solving a class of nonconvex non-concave min-max problem and establish O( −4 ) complexity for finding -first-order stationary point, in which the algorithm only requires invoking one stochastic first-order oracle while enjoying state-of-the-art iteration complexity achieved by stochastic extragradient method by (Iusem et al., 2017). Then we propose an adaptive variant of OSG named Optimistic Adagrad (OAdagrad) and reveal an improved adaptive complexity O − 2 1−α 1 , where α characterizes the growth rate of the cumulative stochastic gradient and 0 ≤ α ≤ 1/2. To the best of our knowledge, this is the first work for establishing adaptive complexity in nonconvex non-concave min-max optimization. Empirically, our experiments show that indeed adaptive gradient algorithms outperform their non-adaptive counterparts in GAN training. Moreover, this observation can be explained by the slow growth rate of the cumulative stochastic gradient, as observed empirically. * Correspondence to mingrui-liu@uiowa.edu 1 Here O(·) compresses a logarithmic factor of .Published as a conference paper at ICLR 2020 2017; Gulrajani et al., 2017). Using Adam is important in GAN training, since replacing it with non-adaptive methods (e.g. SGD) would significantly deteriorate the performance. This paper studies and attempts to answer the following question:Published as a conference paper at ICLR 2020 Vaishnavh Nagarajan and J Zico Kolter. Gradient descent gan optimization is locally stable. In Advances in Neural Information Processing Systems, pp. 5585-5595, 2017. Arkadi Nemirovski. Prox-method with rate of convergence o (1/t) for variational inequalities with lipschitz continuous monotone operators and smooth convex-concave saddle point problems. SIAM Journal on Optimization, 15(1):229-251, 2004.Arkadi Nemirovski and D Yudin. On cezari?s convergence of the steepest descent method for approximating saddle point of convex-concave functions. | TOWARDS BETTER UNDERSTANDING OF ADAPTIVE GRADIENT ALGORITHMS IN GENERATIVE ADVER- SARIAL NETS |
d5212627 | In this work we perform outlier detection using ensembles of neural networks obtained by variational approximation of the posterior in a Bayesian neural network setting. The variational parameters are obtained by sampling from the true posterior by gradient descent. We show our outlier detection results are comparable to those obtained using other efficient ensembling methods. * Equal contribution. | EFFICIENT VARIATIONAL BAYESIAN NEURAL NET- WORK ENSEMBLES FOR OUTLIER DETECTION |
d259212405 | Data heterogeneity is one of the most challenging issues in federated learning, which motivates a variety of approaches to learn personalized models for participating clients. One such approach in deep neural networks based tasks is employing a shared feature representation and learning a customized classifier head for each client. However, previous works do not utilize the global knowledge during local representation learning and also neglect the fine-grained collaboration between local classifier heads, which limit the model generalization ability. In this work, we conduct explicit local-global feature alignment by leveraging global semantic knowledge for learning a better representation. Moreover, we quantify the benefit of classifier combination for each client as a function of the combining weights and derive an optimization problem for estimating optimal weights. Finally, extensive evaluation results on benchmark datasets with various heterogeneous data scenarios demonstrate the effectiveness of our proposed method 1 . . Debiasing model updates for improving personalized federated training. In | Published as a conference paper at ICLR 2023 PERSONALIZED FEDERATED LEARNING WITH FEATURE ALIGNMENT AND CLASSIFIER COLLABORATION |
d257405475 | Causal representation learning is the task of identifying the underlying causal variables and their relations from high-dimensional observations, such as images. Recent work has shown that one can reconstruct the causal variables from temporal sequences of observations under the assumption that there are no instantaneous causal relations between them. In practical applications, however, our measurement or frame rate might be slower than many of the causal effects. This effectively creates "instantaneous" effects and invalidates previous identifiability results. To address this issue, we propose iCITRIS, a causal representation learning method that allows for instantaneous effects in intervened temporal sequences when intervention targets can be observed, e.g., as actions of an agent. iCITRIS identifies the potentially multidimensional causal variables from temporal observations, while simultaneously using a differentiable causal discovery method to learn their causal graph. In experiments on three datasets of interactive systems, iCITRIS accurately identifies the causal variables and their causal graph. * Qualcomm AI Research is an initiative of Qualcomm Technologies, Inc.Published as a conference paper at ICLR 2023 effects, neither of these two tasks can be completed without the other: without knowing the variables, we cannot identify the graph; but without knowing the graph, we cannot identify the causal variables, since they are not conditionally independent. In particular, in contrast to causal relations across time steps, the orientations of instantaneous edges are not determined by the temporal ordering, hence requiring to jointly solve the tasks of causal representation learning and causal discovery.As a starting point, we consider the setting of CITRIS (Causal Identifiability from Temporal Intervened Sequences; Lippe et al.(2022b)). In CITRIS, potentially multidimensional causal variables interact over time, and interventions with known targets may have been performed. While in that work all causal relations were assumed to be temporal, i.e., from variables in one time step to variables in the next time step, we generalize this setting to include instantaneous causal effects. In particular, we show that in general, causal variables are not identifiable if we do not have access to partiallyperfect interventions, i.e., interventions that remove the instantaneous parents. If such interventions are available, we prove that we can identify the minimal causal variables (Lippe et al., 2022b), i.e., the parts of the causal variables that are affected by the interventions, and their temporal and instantaneous causal graph. Our results generalize the identifiability results of Lippe et al. (2022b), since if there are no instantaneous causal relations, any intervention is partially-perfect by definition. As a practical implementation, we propose instantaneous CITRIS (iCITRIS). iCITRIS maps high-dimensional observations, e.g., images, to a lower-dimensional latent space on which it learns an instantaneous causal graph by integrating differentiable causal discovery methods into its prior (Lippe et al., 2022a;Zheng et al., 2018). In experiments on three different video datasets, iCITRIS accurately identifies the causal variables as well as their instantaneous and temporal causal graph. Our contributions are:• We show that causal variables in temporal sequences with instantaneous effects are not identifiable without interventions that remove instantaneous parents. • We prove that when having access to such interventions with known targets, the minimal causal variables can be identified along with their causal graph under mild assumptions. • We propose iCITRIS, a causal representation learning method that identifies minimal causal variables and their causal graph even in the case of instantaneous causal effects. | Published as a conference paper at ICLR 2023 CAUSAL REPRESENTATION LEARNING FOR INSTANTANEOUS AND TEMPORAL EFFECTS IN INTERACTIVE SYSTEMS |
d226226934 | Complex, multi-task problems have proven to be difficult to solve efficiently in a sparse-reward reinforcement learning setting. In order to be sample efficient, multi-task learning requires reuse and sharing of low-level policies. To facilitate the automatic decomposition of hierarchical tasks, we propose the use of step-bystep human demonstrations in the form of natural language instructions and action trajectories. We introduce a dataset of such demonstrations in a crafting-based grid world. Our model consists of a high-level language generator and low-level policy, conditioned on language. We find that human demonstrations help solve the most complex tasks. We also find that incorporating natural language allows the model to generalize to unseen tasks in a zero-shot setting and to learn quickly from a few demonstrations. Generalization is not only reflected in the actions of the agent, but also in the generated natural language instructions in unseen tasks. Our approach also gives our trained agent interpretable behaviors because it is able to generate a sequence of high-level descriptions of its actions. | Published as a conference paper at ICLR 2021 ASK YOUR HUMANS: USING HUMAN INSTRUCTIONS TO IMPROVE GENERALIZATION IN REINFORCEMENT LEARNING |
d338016 | CMF is a technique for simultaneously learning low-rank representations based on a collection of matrices with shared entities. A typical example is the joint modeling of useritem, item-property, and user-feature matrices in a recommender system. The key idea in CMF is that the embeddings are shared across the matrices, which enables transferring information between them. The existing solutions, however, break down when the individual matrices have low-rank structure not shared with others. In this work we present a novel CMF solution that allows each of the matrices to have a separate low-rank structure that is independent of the other matrices, as well as structures that are shared only by a subset of them. We compare MAP and variational Bayesian solutions based on alternating optimization algorithms and show that the model automatically infers the nature of each factor using group-wise sparsity. Our approach supports in a principled way continuous, binary and count observations and is efficient for sparse matrices involving missing data. We illustrate the solution on a number of examples, focusing in particular on an interesting use-case of augmented multi-view learning. | Group-sparse Embeddings in Collective Matrix Factorization |
d231846909 | Adversarial patches pose a realistic threat model for physical world attacks on autonomous systems via their perception component. Autonomous systems in safety-critical domains such as automated driving should thus contain a fail-safe fallback component that combines certifiable robustness against patches with efficient inference while maintaining high performance on clean inputs. We propose BAGCERT, a novel combination of model architecture and certification procedure that allows efficient certification. We derive a loss that enables end-to-end optimization of certified robustness against patches of different sizes and locations. On CIFAR10, BAGCERT certifies 10.000 examples in 43 seconds on a single GPU and obtains 86% clean and 60% certified accuracy against 5 × 5 patches. | Published as a conference paper at ICLR 2021 EFFICIENT CERTIFIED DEFENSES AGAINST PATCH AT- TACKS ON IMAGE CLASSIFIERS |
d3303815 | Planning problems in partially observable environments cannot be solved directly with convolutional networks and require some form of memory. But, even memory networks with sophisticated addressing schemes are unable to learn intelligent reasoning satisfactorily due to the complexity of simultaneously learning to access memory and plan. To mitigate these challenges we propose the Memory Augmented Control Network (MACN). The network splits planning into a hierarchical process. At a lower level, it learns to plan in the locally observed space. At a higher level, it uses a collection of policies computed on locally observed spaces to learn the optimal plan in the global environment it is operating in. The performance of the network is evaluated in discrete grid world environments for path planning in the presence of simple and complex obstacles and in addition is tested for its ability to generalize to new environments not seen in the training set. | Under review as a conference paper at ICLR 2018 MEMORY AUGMENTED CONTROL NETWORKS |
d252693220 | Distilling knowledge from a large teacher model to a lightweight one is a widely successful approach for generating compact, powerful models in the semi-supervised learning setting where a limited amount of labeled data is available. In large-scale applications, however, the teacher tends to provide a large number of incorrect soft-labels that impairs student performance. The sheer size of the teacher additionally constrains the number of soft-labels that can be queried due to prohibitive computational and/or financial costs. The difficulty in achieving simultaneous efficiency (i.e., minimizing soft-label queries) and robustness (i.e., avoiding student inaccuracies due to incorrect labels) hurts the widespread application of knowledge distillation to many modern tasks. In this paper, we present a parameter-free approach with provable guarantees to query the soft-labels of points that are simultaneously informative and correctly labeled by the teacher. At the core of our work lies a game-theoretic formulation that explicitly considers the inherent trade-off between the informativeness and correctness of input instances. We establish bounds on the expected performance of our approach that hold even in worst-case distillation instances. We present empirical evaluations on popular benchmarks that demonstrate the improved distillation performance enabled by our work relative to that of state-of-the-art active learning and active distillation methods. | Published as a conference paper at ICLR 2023 ROBUST ACTIVE DISTILLATION |
d211132970 | Learned joint representations of images and text form the backbone of several important cross-domain tasks such as image captioning. Prior work mostly maps both domains into a common latent representation in a purely supervised fashion. This is rather restrictive, however, as the two domains follow distinct generative processes. Therefore, we propose a novel semi-supervised framework, which models shared information between domains and domain-specific information separately. The information shared between the domains is aligned with an invertible neural network. Our model integrates normalizing flow-based priors for the domain-specific information, which allows us to learn diverse many-to-many mappings between the two domains. We demonstrate the effectiveness of our model on diverse tasks, including image captioning and text-to-image synthesis. | Published as a conference paper at ICLR 2020 LATENT NORMALIZING FLOWS FOR MANY-TO-MANY CROSS-DOMAIN MAPPINGS |
d253098700 | While large language models (LLMs) like GPT-3 have achieved impressive results on multiple choice question answering (MCQA) tasks in the zero, one, and few-shot settings, they generally lag behind the MCQA state of the art (SOTA). MCQA tasks have traditionally been presented to LLMs like cloze tasks. An LLM is conditioned on a question (without the associated answer options) and its chosen option is the one assigned the highest probability after normalization (for length, etc.). A more natural prompting approach is to present the question and answer options to the LLM jointly and have it output the symbol (e.g., "A") associated with its chosen answer option. This approach allows the model to explicitly compare answer options, reduces computational costs, and mitigates the effects of tokenization scheme and answer option representations on answer selection. For the natural approach to be effective, the LLM it is used with must be able to associate answer options with the symbols that represent them. The LLM needs what we term multiple choice symbol binding (MCSB) ability. This ability varies greatly by model. We show that a model with high MCSB ability performs much better with the natural approach than with the traditional approach across 20 diverse datasets and largely closes the gap with the SOTA, suggesting that the MCQA ability of LLMs has been previously underestimated.Published as a conference paper at ICLR 2023 about and compare different candidate answers, and 4) finicky normalization due to tokenization schemes. The centerpiece of our paper is an extensive investigation of an alternative: we explain how these problems might be solved by what we call multiple choice prompting (MCP). In MCP, the language model receives both the question and also a list of candidate answers as on a multiple choice test, with each answer associated with (or "bound" to) a symbol such as "A", "B", "C", etc. We explain how this approach might be why MCP outperforms CP in Section 3.More importantly, though, we demonstrate that when we prompt LLMs with MCP instead of CP, performance often dramatically improves -approaching or even surpassing SOTA performance. On a varied group of 20 datasets, we show that MCP outperforms CP on all but 4 of the datasets, with a mean gap of 9.7% on all tasks and a max gap of 44%. MCP surpasses old SOTA scores on 9 of 20 datasets (by as much as 15% on a single task), and averaged across all datasets, MCP scores fall 0.6% shy of SOTA.This implies that the de facto method for prompting LLMs has led them to be considerably underestimated for MCQA, and that there exists a better general way to prompt a single LLM that scores within a percent of accuracy of all other previous SOTA scores, on average. For the 20 different datasets we consider, SOTA accuracy required 14 customized models and approaches -nearly three individualized setups for every four datasets. We argue that the fact that MCP is comparable to or surpasses SOTA, with no task-specific tuning, is evidence for the efficiency, generality, and overall promise of foundation models in MCQA.Our primary contribution is three-fold: 1) We present an argument for multiple-choice prompting over cloze prompting and formally define multiple choice symbol binding (MCSB), a required ability for an LLM to benefit from MCP; 2) We show that not all LLMs are equally skilled in this regard; and 3) Across 20 diverse datasets, we show that the models most capable of MCSB can individually approach or beat SOTA on most of the considered tasks when prompted with multiple choice prompting instead of the near-universal approach of cloze prompting. Code is available. 1 | Published as a conference paper at ICLR 2023 LEVERAGING LARGE LANGUAGE MODELS FOR MULTIPLE CHOICE QUESTION ANSWERING |
d53158404 | This paper addresses the problem of incremental domain adaptation (IDA) in natural language processing (NLP). We assume each domain comes one after another, and that we could only access data in the current domain. The goal of IDA is to build a unified model performing well on all the domains that we have encountered. We adopt the recurrent neural network (RNN) widely used in NLP, but augment it with a directly parameterized memory bank, which is retrieved by an attention mechanism at each step of RNN transition. The memory bank provides a natural way of IDA: when adapting our model to a new domain, we progressively add new slots to the memory bank, which increases the number of parameters, and thus the model capacity. We learn the new memory slots and fine-tune existing parameters by back-propagation. Experimental results show that our approach achieves significantly better performance than fine-tuning alone. Compared with expanding hidden states, our approach is more robust for old domains, shown by both empirical and theoretical results. Our model also outperforms previous work of IDA including elastic weight consolidation and progressive neural networks in the experiments. 1 * Equal contribution.Published as a conference paper at ICLR 2020 few steps of gradient updates does transfer quickly, but it suffers from the catastrophic forgetting problem(Kirkpatrick et al., 2017). Suppose during prediction a data point is not labeled with its domain, the (single) fine-tuned model cannot predict well for samples in previous domains, as it tends to "forget" quickly during fine-tuning.A recent trend of domain adaptation in the deep learning regime is the progressive neural network(Rusu et al., 2016), which progressively grows the network capacity if a new domain comes. Typically, this is done by enlarging the model with new hidden states and a new predictor(Figure 1a). To avoid interfering with existing knowledge, the newly added hidden states are not fed back to the previously trained states. During training, all existing parameters are frozen, and only the newly added ones are trained. For inference, the new predictor is used for all domains, which is sometimes undesired as the new predictor is trained with only the last domain. | Published as a conference paper at ICLR 2020 PROGRESSIVE MEMORY BANKS FOR INCREMENTAL DOMAIN ADAPTATION |
d9864100 | Learning a natural language interface for database tables is a challenging task that involves deep language understanding and multi-step reasoning. The task is often approached by mapping natural language queries to logical forms or programs that provide the desired response when executed on the database. To our knowledge, this paper presents the first weakly supervised, end-to-end neural network model to induce such programs on a real-world dataset. We enhance the objective function of Neural Programmer, a neural network with built-in discrete operations, and apply it on WikiTableQuestions, a natural language question-answering dataset. The model is trained end-to-end with weak supervision of question-answer pairs, and does not require domain-specific grammars, rules, or annotations that are key elements in previous approaches to program induction. The main experimental result in this paper is that a single Neural Programmer model achieves 34.2% accuracy using only 10,000 examples with weak supervision. An ensemble of 15 models, with a trivial combination technique, achieves 37.7% accuracy, which is competitive to the current state-of-the-art accuracy of 37.1% obtained by a traditional natural language semantic parser.Recently, many neural network models have been developed for program induction (Andreas et al., | Published as a conference paper at ICLR 2017 LEARNING A NATURAL LANGUAGE INTERFACE WITH NEURAL PROGRAMMER |
d2684987 | We introduce a parametric nonlinear transformation that is well-suited for Gaussianizing data from natural images. The data are linearly transformed, and each component is then normalized by a pooled activity measure, computed by exponentiating a weighted sum of rectified and exponentiated components and a constant. We optimize the parameters of the full transformation (linear transform, exponents, weights, constant) over a database of natural images, directly minimizing the negentropy of the responses. The optimized transformation substantially Gaussianizes the data, achieving a significantly smaller mutual information between transformed components than alternative methods including ICA and radial Gaussianization. The transformation is differentiable and can be efficiently inverted, and thus induces a density model on images. We show that samples of this model are visually similar to samples of natural image patches. We demonstrate the use of the model as a prior probability density that can be used to remove additive noise. Finally, we show that the transformation can be cascaded, with each layer optimized using the same Gaussianization objective, thus offering an unsupervised method of optimizing a deep network architecture. | DENSITY MODELING OF IMAGES USING A GENERALIZED NORMALIZATION TRANSFORMATION |
d14480911 | Optimization by stochastic gradient descent is an important component of many large-scale machine learning algorithms. A wide variety of such optimization algorithms have been devised; however, it is unclear whether these algorithms are robust and widely applicable across many different optimization landscapes. In this paper we develop a collection of unit tests for stochastic optimization. Each unit test rapidly evaluates an optimization algorithm on a small-scale, isolated, and well-understood difficulty, rather than in real-world scenarios where many such issues are entangled. Passing these unit tests is not sufficient, but absolutely necessary for any algorithms with claims to generality or robustness. We give initial quantitative and qualitative results on numerous established algorithms. The testing framework is open-source, extensible, and easy to apply to new algorithms. | Unit Tests for Stochastic Optimization |
d231632202 | Learning from a limited number of samples is challenging since the learned model can easily become overfitted based on the biased distribution formed by only a few training examples. In this paper, we calibrate the distribution of these fewsample classes by transferring statistics from the classes with sufficient examples. Then an adequate number of examples can be sampled from the calibrated distribution to expand the inputs to the classifier. We assume every dimension in the feature representation follows a Gaussian distribution so that the mean and the variance of the distribution can borrow from that of similar classes whose statistics are better estimated with an adequate number of samples. Our method can be built on top of off-the-shelf pretrained feature extractors and classification models without extra parameters. We show that a simple logistic regression classifier trained using the features sampled from our calibrated distribution can outperform the state-of-the-art accuracy on three datasets (5% improvement on miniImageNet compared to the next best). The visualization of these generated features demonstrates that our calibrated distribution is an accurate estimation. | Published as a conference paper at ICLR 2021 FREE LUNCH FOR FEW-SHOT LEARNING: DISTRIBUTION CALIBRATION |
d2428314 | Categorical variables are a natural choice for representing discrete structure in the world. However, stochastic neural networks rarely use categorical latent variables due to the inability to backpropagate through samples. In this work, we present an efficient gradient estimator that replaces the non-differentiable sample from a categorical distribution with a differentiable sample from a novel Gumbel-Softmax distribution. This distribution has the essential property that it can be smoothly annealed into a categorical distribution. We show that our Gumbel-Softmax estimator outperforms state-of-the-art gradient estimators on structured output prediction and unsupervised generative modeling tasks with categorical latent variables, and enables large speedups on semi-supervised classification. * Work done during an internship at Google Brain. | Published as a conference paper at ICLR 2017 CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX |
d235765481 | We propose to address quadrupedal locomotion tasks using Reinforcement Learning (RL) with a Transformer-based model that learns to combine proprioceptive information and high-dimensional depth sensor inputs. While learning-based locomotion has made great advances using RL, most methods still rely on domain randomization for training blind agents that generalize to challenging terrains.Our key insight is that proprioceptive states only offer contact measurements for immediate reaction, whereas an agent equipped with visual sensory observations can learn to proactively maneuver environments with obstacles and uneven terrain by anticipating changes in the environment many steps ahead. In this paper, we introduce LocoTransformer, an end-to-end RL method that leverages both proprioceptive states and visual observations for locomotion control. We evaluate our method in challenging simulated environments with different obstacles and uneven terrain. We transfer our learned policy from simulation to a real robot by running it indoors and in the wild with unseen obstacles and terrain. Our method not only significantly improves over baselines, but also achieves far better generalization performance, especially when transferred to the real robot. Our project page with videos is at https | LEARNING VISION-GUIDED QUADRUPEDAL LOCO- MOTION END-TO-END WITH CROSS-MODAL TRANS- FORMERS |
d224470441 | Recently, Neural Topic Models (NTMs) inspired by variational autoencoders have obtained increasingly research interest due to their promising results on text analysis. However, it is usually hard for existing NTMs to achieve good document representation and coherent/diverse topics at the same time. Moreover, they often degrade their performance severely on short documents. The requirement of reparameterisation could also comprise their training quality and model flexibility.To address these shortcomings, we present a new neural topic model via the theory of optimal transport (OT). Specifically, we propose to learn the topic distribution of a document by directly minimising its OT distance to the document's word distributions. Importantly, the cost matrix of the OT distance models the weights between topics and words, which is constructed by the distances between topics and words in an embedding space. Our proposed model can be trained efficiently with a differentiable loss. Extensive experiments show that our framework significantly outperforms the state-of-the-art NTMs on discovering more coherent and diverse topics and deriving better document representations for both regular and short texts. . Improving topic models with latent feature word representations. TACL, 3:299-313, 2015. | Published as a conference paper at ICLR 2021 NEURAL TOPIC MODEL VIA OPTIMAL TRANSPORT |
d246430569 | Our work focuses on the development of a learnable neural representation of human pose for advanced AI assisted animation tooling. Specifically, we tackle the problem of constructing a full static human pose based on sparse and variable user inputs (e.g. locations and/or orientations of a subset of body joints). To solve this problem, we propose a novel neural architecture that combines residual connections with prototype encoding of a partially specified pose to create a new complete pose from the learned latent space. We show that our architecture outperforms a baseline based on Transformer, both in terms of accuracy and computational efficiency. Additionally, we develop a user interface to integrate our neural model in Unity, a real-time 3D development platform. Furthermore, we introduce two new datasets representing the static human pose modeling problem, based on high-quality human motion capture data. Our code is publically available here: https://github.com/boreshkinai/protores. | Published as a conference paper at ICLR 2022 PROTORES: PROTO-RESIDUAL NETWORK FOR POSE AUTHORING VIA LEARNED INVERSE KINEMATICS |
d204960946 | Translation into morphologically-rich languages challenges neural machine translation (NMT) models with extremely sparse vocabularies where atomic treatment of surface forms is unrealistic. This problem is typically addressed by either pre-processing words into subword units or performing translation directly at the level of characters. The former is based on word segmentation algorithms optimized using corpus-level statistics with no regard to the translation task. The latter learns directly from translation data but requires rather deep architectures. In this paper, we propose to translate words by modeling word formation through a hierarchical latent variable model which mimics the process of morphological inflection. Our model generates words one character at a time by composing two latent representations: a continuous one, aimed at capturing the lexical semantics, and a set of (approximately) discrete features, aimed at capturing the morphosyntactic function, which are shared among different surface forms. Our model achieves better accuracy in translation into three morphologically-rich languages than conventional open-vocabulary NMT methods, while also demonstrating a better generalization capacity under low to mid-resource settings. | A LATENT MORPHOLOGY MODEL FOR OPEN- VOCABULARY NEURAL MACHINE TRANSLATION |
d258833259 | Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like Horn clauses such as R 1 (x, z) ∧ R 2 (z, y) ⇒ H(x, y). This formulation overlooks additional contextual information from neighboring sub-graphs of entity variables x, y and z. Intuitively, there is a large gap here, as local sub-graphs have been found to provide important information for knowledge graph completion. Inspired by these observations, we propose Logical Entity RePresentation (LERP) to encode contextual information of entities in the knowledge graph. A LERP is designed as a vector of probabilistic logical functions on the entity's neighboring sub-graph. It is an interpretable representation while allowing for differentiable optimization. We can then incorporate LERP into probabilistic logical rule learning to learn more expressive rules. Empirical results demonstrate that with LERP, our model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods. Moreover, we find that our model can discover a more expressive family of logical rules. LERP can also be further combined with embedding learning methods like TransE to make it more interpretable. 1 | LOGICAL ENTITY REPRESENTATION IN KNOWLEDGE- GRAPHS FOR DIFFERENTIABLE RULE LEARNING |
d30043042 | A generative model is developed for deep (multi-layered) convolutional dictionary learning. A novel probabilistic pooling operation is integrated into the deep model, yielding efficient bottom-up (pretraining) and top-down (refinement) probabilistic learning. Experimental results demonstrate powerful capabilities of the model to learn multi-layer features from images, and excellent classification results are obtained on the MNIST and Caltech 101 datasets. | A GENERATIVE MODEL FOR DEEP CONVOLUTIONAL LEARNING |
d254018220 | The rise of generalist large-scale models in natural language and vision has made us expect that a massive data-driven approach could achieve broader generalization in other domains such as continuous control. In this work, we explore a method for learning a single policy that manipulates various forms of agents to solve various tasks by distilling a large amount of proficient behavioral data. In order to align input-output (IO) interface among multiple tasks and diverse agent morphologies while preserving essential 3D geometric relations, we introduce morphology-task graph, which treats observations, actions and goals/task in a unified graph representation. We also develop MxT-Bench for fast large-scale behavior generation, which supports procedural generation of diverse morphology-task combinations with a minimal blueprint and hardware-accelerated simulator. Through efficient representation and architecture selection on MxT-Bench, we find out that a morphology-task graph representation coupled with Transformer architecture improves the multi-task performances compared to other baselines including recent discrete tokenization, and provides better prior knowledge for zero-shot transfer or sample efficiency in downstream multi-task imitation learning. Our work suggests large diverse offline datasets, unified IO representation, and policy representation and architecture selection through supervised learning form a promising approach for studying and advancing morphology-task generalization 1 . * Work done as Student Researcher at Google. . Flamingo: a visual language model for few-shot learning. arXiv preprint arxiv:. Towards human-level bimanual dexterous manipulation with reinforcement learning. arXiv preprint arXiv:2206.08686, 2022b. Sutskever. Learning transferable visual models from natural language supervision. arXiv preprint arXiv:2103.00020, 2021. | A SYSTEM FOR MORPHOLOGY-TASK GENERALIZA- TION VIA UNIFIED REPRESENTATION AND BEHAVIOR DISTILLATION |
d209501080 | In this work we study locality and compositionality in the context of learning representations for Zero Shot Learning (ZSL). In order to well-isolate the importance of these properties in learned representations, we impose the additional constraint that, differently from most recent work in ZSL, no pre-training on different datasets (e.g. ImageNet) is performed. The results of our experiments show how locality, in terms of small parts of the input, and compositionality, i.e. how well can the learned representations be expressed as a function of a smaller vocabulary, are both deeply related to generalization and motivate the focus on more local-aware models in future research directions for representation learning. | Published as a conference paper at ICLR 2020 LOCALITY AND COMPOSITIONALITY IN ZERO-SHOT LEARNING |
d231632629 | Recently proposed consistency-based Semi-Supervised Learning (SSL) methods such as the Π-model, temporal ensembling, the mean teacher, or the virtual adversarial training, have advanced the state of the art in several SSL tasks. These methods can typically reach performances that are comparable to their fully supervised counterparts while using only a fraction of labelled examples. Despite these methodological advances, the understanding of these methods is still relatively limited. In this text, we analyse (variations of) the Π-model in settings where analytically tractable results can be obtained. We establish links with Manifold Tangent Classifiers and demonstrate that the quality of the perturbations is key to obtaining reasonable SSL performances. Importantly, we propose a simple extension of the Hidden Manifold Model that naturally incorporates data-augmentation schemes and offers a framework for understanding and experimenting with SSL methods.arXiv:2101.06967v1 [stat.ML] 18 Jan 2021Published as a conference paper at ICLR 2021Contributions: consistency-based semi-supervised learning methods have recently been shown to achieve state-of-the-art results. Despite these methodological advances, the understanding of these methods is still relatively limited when compared to the fully-supervised setting (Saxe et al.In this article, we do not propose a new SSL method. Instead, we analyse consistency-based methods in settings where analytically tractable results can be obtained, when the data-samples lie in the neighbourhood of well-defined and tractable low-dimensional manifolds, and simple and controlled experiments can be carried out. We establish links with Manifold Tangent Classifiers and demonstrate that consistency-based SSL methods are in general more powerful since they can better exploit the local geometry of the data-manifold if efficient data-augmentation/perturbation schemes are used. Furthermore, in section 4.1 we show that the popular Mean Teacher method and the conceptually more simple Π-model approach share the same solutions in the regime when the data-augmentations are small; this confirms often reported claim that the data-augmentation schemes leveraged by the recent SSL, as well as fully unsupervised algorithms, are instrumental to their success. Finally, in section 4.3 we propose an extension of the Hidden Manifold Model(Goldt et al., 2019;Gerace et al., 2020). This generative model allows us to investigate the properties of consistency-based SSL methods, taking into account the data-augmentation process and the underlying low-dimensionality of the data, in a simple and principled manner, and without relying on a specific dataset. For gaining understanding of SSL, as well as self-supervised learning methods, we believe it to be important to develop a framework that (i) can take into account the geometry of the data (ii) allows the study of the influence of the quality of the data-augmentation schemes (iii) does not rely on any particular dataset. While the understanding of fully-supervised methods have largely been driven by the analysis of simplified model architectures (eg. linear and two-layered models, large dimension asymptotic such as the Neural Tangent Kernel), these analytical tools alone are unlikely to be enough to explain the mechanisms responsible for the success of SSL and self-supervised learning methods Chen et al. (2020); Grill et al. (2020), since they do not, and cannot easily be extended to, account for the geometry of the data and data-augmentation schemes. Our proposed framework offers a small step in that direction. | ON DATA-AUGMENTATION AND CONSISTENCY- BASED SEMI-SUPERVISED LEARNING |
d250425754 | In this paper, we show that recent advances in self-supervised representation learning enable unsupervised object discovery and semantic segmentation with a performance that matches the state of the field on supervised semantic segmentation 10 years ago. We propose a methodology based on unsupervised saliency masks and self-supervised feature clustering to kickstart object discovery followed by training a semantic segmentation network on pseudo-labels to bootstrap the system on images with multiple objects. We show that while being conceptually simple our proposed baseline is surprisingly strong. We present results on PASCAL VOC that go far beyond the current state of the art (50.0 mIoU) , and we report for the first time results on MS COCO for the whole set of 81 classes: our method discovers 34 categories with more than 20% IoU, while obtaining an average IoU of 19.6 for all 81 categories. Figure 1: Unsupervised semantic segmentation predictions on PASCAL VOC (Everingham et al., 2012). Our COMUS does not use human annotations to discover objects and their precise localization. In contrast to the prior state-of-the-art method MaskContrast (Van Gansbeke et al., 2021), COMUS yields more precise segmentations, avoids confusion of categories, and is not restricted to only one object category per image. | Published as a conference paper at ICLR 2023 UNSUPERVISED SEMANTIC SEGMENTATION WITH SELF-SUPERVISED OBJECT-CENTRIC REPRESENTA- TIONS |
d26100519 | We present Deep Voice 3, a fully-convolutional attention-based neural textto-speech (TTS) system. Deep Voice 3 matches state-of-the-art neural speech synthesis systems in naturalness while training ten times faster. We scale Deep Voice 3 to data set sizes unprecedented for TTS, training on more than eight hundred hours of audio from over two thousand speakers. In addition, we identify common error modes of attention-based speech synthesis networks, demonstrate how to mitigate them, and compare several different waveform synthesis methods. We also describe how to scale inference to ten million queries per day on one single-GPU server.2. We show that our architecture trains quickly and scales to the LibriSpeech dataset(Panayotov et al., 2015), which consists of nearly 820 hours of audio data from 2484 speakers.3. We demonstrate that we can generate monotonic attention behavior, avoiding error modes commonly occurred in speech synthesis. 4. We compare the quality of several waveform synthesis methods for a single speaker, including WORLD (Morise et al., 2016), Griffin-Lim (Griffin & Lim, 1984), and WaveNet (Oord et al., 2016). 5. We describe the implementation of an inference kernel for Deep Voice 3, which can serve up to ten million queries per day on one single-GPU server. * Authors listed in reverse alphabetical order. † These authors contributed to this work while members of Baidu Research. Under review as a conference paper at ICLR 2018 2 RELATED WORK Our work builds upon the state-of-the-art in neural speech synthesis and attention-based sequenceto-sequence learning. Several recent works tackle the problem of synthesizing speech with neural networks, including Deep Voice 1 (Arık et al., 2017), Deep Voice 2 (Arık et al., 2017), Tacotron (Wang et al., 2017), Char2Wav (Sotelo et al., 2017), VoiceLoop (Taigman et al., 2017), SampleRNN (Mehri et al., 2017), and WaveNet (Oord et al., 2016). Deep Voice 1 & 2 retain the traditional structure of TTS pipelines, separating grapheme-to-phoneme conversion, duration and frequency prediction, and waveform synthesis. In contrast to Deep Voice 1 & 2, Deep Voice 3 employs an attention-based sequenceto-sequence model, yielding a more compact architecture. Similar to Deep Voice 3, Tacotron and Char2Wav are the two proposed sequence-to-sequence models for neural TTS. Tacotron is a neural text-to-spectrogram conversion model, used with Griffin-Lim for spectrogram-to-waveform synthesis. Char2Wav predicts the parameters of WORLD vocoder (Morise et al., 2016) and uses a Sam-pleRNN conditioned upon WORLD parameters for waveform generation. In contrast to Char2Wav and Tacotron, Deep Voice 3 avoids Recurrent Neural Networks (RNNs) 1 to speed up training and alleviates several challenging error modes that attention models fall into. Thus, Deep Voice 3 makes attention-based TTS feasible for a production TTS system with no compromise on accuracy. Finally, WaveNet and SampleRNN are proposed as neural vocoder models for waveform synthesis. It is also worth noting that there are numerous alternatives for high-quality hand-engineered vocoders in the literature, such as STRAIGHT(Kawahara et al., 1999), Vocaine (Agiomyrgiannakis, 2015), and WORLD (Morise et al., 2016). Deep Voice 3 adds no novel vocoder, but has the potential to be integrated with different waveform synthesis methods with slight modifications of its architecture. | DEEP VOICE 3: 2000-SPEAKER NEURAL TEXT-TO-SPEECH |
d16165021 | We develop a new statistical model for photographic images, in which the local responses of a bank of linear filters are described as jointly Gaussian, with zero mean and a covariance that varies slowly over spatial position. We optimize sets of filters so as to minimize the nuclear norm of matrices of their local activations (i.e., the sum of the singular values), thus encouraging a flexible form of sparsity that is not tied to any particular dictionary or coordinate system. Filters optimized according to this objective are oriented and band-pass, and their responses exhibit substantial local correlation. We show that images can be reconstructed nearly perfectly from estimates of the local filter response covariance alone, and with minimal degradation (either visual or MSE) from low-rank approximations of these covariances. As such, this representation holds much promise for use in applications such as denoising, compression, and texture representation, and may form a useful substrate for hierarchical decompositions. | THE LOCAL LOW-DIMENSIONALITY OF NATURAL IMAGES |
d231847094 | Sparsity in Deep Neural Networks (DNNs) has been widely studied to compress and accelerate the models on resource-constrained environments. It can be generally categorized into unstructured fine-grained sparsity that zeroes out multiple individual weights distributed across the neural network, and structured coarsegrained sparsity which prunes blocks of sub-networks of a neural network. Finegrained sparsity can achieve a high compression ratio but is not hardware friendly and hence receives limited speed gains. On the other hand, coarse-grained sparsity cannot concurrently achieve both apparent acceleration on modern GPUs and decent performance. In this paper, we are the first to study training from scratch an N:M fine-grained structured sparse network, which can maintain the advantages of both unstructured fine-grained sparsity and structured coarse-grained sparsity simultaneously on specifically designed GPUs. Specifically, a 2 : 4 sparse network could achieve 2× speed-up without performance drop on Nvidia A100 GPUs. Furthermore, we propose a novel and effective ingredient, sparse-refined straightthrough estimator (SR-STE), to alleviate the negative influence of the approximated gradients computed by vanilla STE during optimization. We also define a metric, Sparse Architecture Divergence (SAD), to measure the sparse network's topology change during the training process. Finally, We justify SR-STE's advantages with SAD and demonstrate the effectiveness of SR-STE by performing comprehensive experiments on various tasks. Source codes and models are available at https://github.com/NM-sparsity/NM-sparsity. | Published as a conference paper at ICLR 2021 LEARNING N:M FINE-GRAINED STRUCTURED SPARSE NEURAL NETWORKS FROM SCRATCH |
d248863441 | Many policy gradient methods are variants of Actor-Critic (AC), where a value function (critic) is learned to facilitate updating the parameterized policy (actor). The update to the actor involves a log-likelihood update weighted by the actionvalues, with the addition of entropy regularization for soft variants. In this work, we explore an alternative update for the actor, based on an extension of the cross entropy method (CEM) to condition on inputs (states). The idea is to start with a broader policy and slowly concentrate around maximally valued actions, using a maximum likelihood update towards actions in the top percentile per state. The speed of this concentration is controlled by a proposal policy, that concentrates at a slower rate than the actor. We first provide a policy improvement result in an idealized setting, and then prove that our conditional CEM (CCEM) strategy tracks a CEM update per state, even with changing action-values. We empirically show that our GreedyAC algorithm, that uses CCEM for the actor update, performs better than Soft Actor-Critic and is much less sensitive to entropy-regularization.arXiv:1810.09103v4 [cs.LG] 28 Feb 2023Published as a conference paper at ICLR 2023In this work we propose a new greedification strategy towards this goal. The basic idea is to iteratively take the top percentile of actions, ranked according to the learned action-values. The procedure slowly concentrates on the maximal action(s), across states, for the given action-values. The update itself is simple: N ∈ N actions are sampled according to a proposal policy, the actions are sorted based on the magnitude of the action-values, and the policy is updated to increase the probability of the ρN maximally valued actions for ρ ∈ (0, 1). We call this algorithm for the actor Conditional CEM (CCEM), because it is an extension of the well-known Cross-Entropy Method (CEM)(Rubinstein, 1999)to condition on inputs 1 . We leverage theory for CEM to validate that our algorithm concentrates on maximally valued actions across states over time. We introduce GreedyAC, a new AC algorithm that uses CCEM for the actor.GreedyAC has several advantages over using Boltzmann greedification. First, we show that our new greedification operator ensures a policy improvement for the original MDP, rather than a different entropy-regularized MDP. Second, we can still leverage entropy to prevent policy collapse, but only incorporate it into the proposal policy. This ensures the agent considers potentially optimal actions for longer, but does not skew the actor. In fact, it is possible to decouple the role of entropy for exploration and policy collapse within GreedyAC: the actor could have a small amount of entropy to encourage exploration, and the proposal policy a higher level of entropy to avoid policy collapse. | GREEDY ACTOR-CRITIC: A NEW CONDITIONAL CROSS-ENTROPY METHOD FOR POLICY IMPROVE- MENT |
d257663312 | Latent confounding has been a long-standing obstacle for causal reasoning from observational data. One popular approach is to model the data using acyclic directed mixed graphs (ADMGs), which describe ancestral relations between variables using directed and bidirected edges. However, existing methods using AD-MGs are based on either linear functional assumptions or a discrete search that is complicated to use and lacks computational tractability for large datasets. In this work, we further extend the existing body of work and develop a novel gradientbased approach to learning an ADMG with non-linear functional relations from observational data. We first show that the presence of latent confounding is identifiable under the assumptions of bow-free ADMGs with non-linear additive noise models. With this insight, we propose a novel neural causal model based on autoregressive flows for ADMG learning. This not only enables us to determine complex causal structural relationships behind the data in the presence of latent confounding, but also estimate their functional relationships (hence treatment effects) simultaneously. We further validate our approach via experiments on both synthetic and real-world datasets, and demonstrate the competitive performance against relevant baselines. | Published as a conference paper at ICLR 2023 CAUSAL REASONING IN THE PRESENCE OF LATENT CONFOUNDERS VIA NEURAL ADMG LEARNING |
d211146346 | Most existing 3D CNNs for video representation learning are clip-based methods, and thus do not consider video-level temporal evolution of spatio-temporal features. In this paper, we propose Video-level 4D Convolutional Neural Networks, referred as V4D, to model the evolution of long-range spatio-temporal representation with 4D convolutions, and at the same time, to preserve strong 3D spatio-temporal representation with residual connections. Specifically, we design a new 4D residual block able to capture inter-clip interactions, which could enhance the representation power of the original clip-level 3D CNNs. The 4D residual blocks can be easily integrated into the existing 3D CNNs to perform long-range modeling hierarchically. We further introduce the training and inference methods for the proposed V4D. Extensive experiments are conducted on three video recognition benchmarks, where V4D achieves excellent results, surpassing recent 3D CNNs by a large margin. | V4D:4D CONVOLUTIONAL NEURAL NETWORKS FOR VIDEO-LEVEL REPRESENTATION LEARNING |
d232428253 | Off-policy evaluation (OPE) holds the promise of being able to leverage large, offline datasets for both evaluating and selecting complex policies for decision making. The ability to learn offline is particularly important in many real-world domains, such as in healthcare, recommender systems, or robotics, where online data collection is an expensive and potentially dangerous process. Being able to accurately evaluate and select high-performing policies without requiring online interaction could yield significant benefits in safety, time, and cost for these applications. While many OPE methods have been proposed in recent years, comparing results between papers is difficult because currently there is a lack of a comprehensive and unified benchmark, and measuring algorithmic progress has been challenging due to the lack of difficult evaluation tasks. In order to address this gap, we present a collection of policies that in conjunction with existing offline datasets can be used for benchmarking off-policy evaluation. Our tasks include a range of challenging high-dimensional continuous control problems, with wide selections of datasets and policies for performing policy selection. The goal of our benchmark is to provide a standardized measure of progress that is motivated from a set of principles designed to challenge and test the limits of existing OPE methods. We perform an evaluation of state-of-the-art algorithms and provide open-source access to our data and code to foster future research in this area † .Published as a conference paper at ICLR 2021 commonly explored by modern deep reinforcement learning algorithms(Bellemare et al., 2013;Brockman et al., 2016) with standardized evaluation protocols and metrics. Our goal is to provide a set of tasks with a range of difficulty, excercise a variety of design properties, and provide policies with different behavioral patterns in order to establish a standardized framework for comparing OPE algorithms. We put particular emphasis on large datasets, long-horizon tasks, and task complexity to facilitate the development of scalable algorithms that can solve high-dimensional problems.Our primary contribution is the Deep Off-Policy Evaluation (DOPE) benchmark. DOPE is designed to measure the performance of OPE methods by 1) evaluating on challenging control tasks with properties known to be difficult for OPE methods, but which occur in real-world scenarios, 2) evaluating across a range of policies with different values, to directly measure performance on policy evaluation, ranking and selection, and 3) evaluating in ideal and adversarial settings in terms of dataset coverage and support. These factors are independent of task difficulty, but are known to have a large impact on OPE performance. To achieve 1, we selected tasks on a set of design principles outlined in Section 3.1. To achieve 2, for each task we include 10 to 96 policies for evaluation and devise an evaluation protocol that measures policy evaluation, ranking, and selection as outlined in Section 3.2. To achieve 3, we provide two domains with differing dataset coverage and support properties described in Section 4. Finally, to enable an easy-to-use research platform, we provide the datasets, target policies, evaluation API, as well as the recorded results of state-of-the-art algorithms (presented in Section 5) as open-source. . Counterfactual reasoning and learning systems: The example of computational advertising. . Deep reinforcement learning in a handful of trials using probabilistic dynamics models. , et al. Reinforcement learning benchmarks and bake-offs ii. | Published as a conference paper at ICLR 2021 BENCHMARKS FOR DEEP OFF-POLICY EVALUATION |
d14025106 | We describe a mechanism for subsampling sequences and show how to compute its expected output so that it can be trained with standard backpropagation. We test this approach on a simple toy problem and discuss its shortcomings.SUBSAMPLING SEQUENCESConsider a mechanism which, given a sequence of vectors s = {s 0 , s 1 , . . . , s T −1 }, s t ∈ R d , produces a sequence of "sampling probabilities" e = {e 0 , e 1 , . . . , e T −1 }, e t ∈ [0, 1] which denote the probability of including s t in the output sequence y = {y 0 , y 1 , . . . , y U −1 }. Producing y from s and e is encapsulated by the following pseudo-code: # Initialize y as an empty sequence y = [] for t in {0, 1, ..., T -1}: # Draw a random number in [0, 1] and compare to e[t] if rand() < e[t]: # Add s[t] to y with probability e[t] y.append(s[t]) * Work done as members of the Google Brain Residency program 1 arXiv:1702.06914v2 [cs.LG] | Workshop track -ICLR 2017 TRAINING A SUBSAMPLING MECHANISM IN EXPECTATION |
d257365796 | Reparameterization aims to improve the generalization of deep neural networks by transforming convolutional layers into equivalent multi-branched structures during training. However, there exists a gap in understanding how reparameterization may change and benefit the learning process of neural networks. In this paper, we present a novel spatial gradient scaling method to redistribute learning focus among weights in convolutional networks. We prove that spatial gradient scaling achieves the same learning dynamics as a branched reparameterization yet without introducing structural changes into the network. We further propose an analytical approach that dynamically learns scalings for each convolutional layer based on the spatial characteristics of its input feature map gauged by mutual information. Experiments on CIFAR-10, CIFAR-100, and ImageNet show that without searching for reparameterized structures, our proposed scaling method outperforms the state-of-the-art reparameterization strategies at a lower computational cost. The | Published as a conference paper at ICLR 2023 REPARAMETERIZATION THROUGH SPATIAL GRADIENT SCALING |
d211838009 | The classical development of neural networks has been primarily for mappings between a finite-dimensional Euclidean space and a set of classes, or between two finite-dimensional Euclidean spaces. The purpose of this work is to generalize neural networks so that they can learn mappings between infinite-dimensional spaces (operators).The key innovation in our work is that a single set of network parameters, within a carefully designed network architecture, may be used to describe mappings between infinite-dimensional spaces and between different finitedimensional approximations of those spaces. We formulate approximation of the infinite-dimensional mapping by composing nonlinear activation functions and a class of integral operators. The kernel integration is computed by message passing on graph networks. This approach has substantial practical consequences which we will illustrate in the context of mappings between input data to partial differential equations (PDEs) and their solutions. In this context, such learned networks can generalize among different approximation methods for the PDE (such as finite difference or finite element methods) and among approximations corresponding to different underlying levels of resolution and discretization. Experiments confirm that the proposed graph kernel network does have the desired properties and show competitive performance compared to the state of the art solvers. | Neural Operator: Graph Kernel Network for Partial Differential Equations |
d235614296 | We aim to help users communicate their intent to machines using flexible, adaptive interfaces that translate arbitrary user input into desired actions. In this work, we focus on assistive typing applications in which a user cannot operate a keyboard, but can instead supply other inputs, such as webcam images that capture eye gaze or neural activity measured by a brain implant. Standard methods train a model on a fixed dataset of user inputs, then deploy a static interface that does not learn from its mistakes; in part, because extracting an error signal from user behavior can be challenging. We investigate a simple idea that would enable such interfaces to improve over time, with minimal additional effort from the user: online learning from user feedback on the accuracy of the interface's actions. In the typing domain, we leverage backspaces as feedback that the interface did not perform the desired action. We propose an algorithm called x-to-text (X2T) that trains a predictive model of this feedback signal, and uses this model to fine-tune any existing, default interface for translating user input into actions that select words or characters. We evaluate X2T through a small-scale online user study with 12 participants who type sentences by gazing at their desired words, a large-scale observational study on handwriting samples from 60 users, and a pilot study with one participant using an electrocorticography-based brain-computer interface. The results show that X2T learns to outperform a non-adaptive default interface, stimulates user co-adaptation to the interface, personalizes the interface to individual users, and can leverage offline data collected from the default interface to improve its initial performance and accelerate online learning. User sees list of words and system's current prediction of desired word ECoG brain-computer interface records user's neural activity Reward model Predicted word Backspace or continue Default interface Our interface Reward User input Train reward model Input-action-reward log dataBinary feedback (e.g., from button press) Anonymized user Anonymized userFigure 1: We formulate assistive typing as a human-in-the-loop decision-making problem, in which the interface observes user inputs (e.g., neural activity measured by a brain implant) and performs actions (e.g., word selections) on behalf of the user. We treat a backspace as feedback from the user that the interface performed the wrong action. By training a model online to predict backspaces, we continually improve the interface.action in response to a given input. By learning from this naturally-occurring feedback signal instead of an explicit label, we do not require any additional effort from the user to improve the interface. Furthermore, because our method is applied on top of the user's default interface, our approach is complementary to other work that develops state-of-the-art, domain-specific methods for problems like gaze tracking and handwriting recognition.Figure 1describes our algorithm: we initialize our model using offline data generated by the default interface, deploy our interface as an augmentation to the default interface, collect online feedback, and update our model.We formulate assistive typing as an online decision-making problem, in which the interface receives observations of user inputs, performs actions that select words or characters, and receives a reward signal that is automatically constructed from the user's backspaces. To improve the default interface's actions, we fit a neural network reward model that predicts the reward signal given the user's input and the interface's action. Upon observing a user input, our interface uses the trained reward model to update the prior policy given by the default interface to a posterior policy conditioned on optimality, then samples an action from this posterior (seeFigure 1). We call this method x-to-text (X2T), where x refers to the arbitrary type of user input; e.g., eye gaze or brain activity. | Published as a conference paper at ICLR 2021 X2T: TRAINING AN X-TO-TEXT TYPING INTERFACE WITH ONLINE LEARNING FROM USER FEEDBACK |
d250280065 | Reliable generalization lies at the heart of safe ML and AI. However, understanding when and how neural networks generalize remains one of the most important unsolved problems in the field. In this work, we conduct an extensive empirical study (20 910 models, 15 tasks) to investigate whether insights from the theory of computation can predict the limits of neural network generalization in practice. We demonstrate that grouping tasks according to the Chomsky hierarchy allows us to forecast whether certain architectures will be able to generalize to out-ofdistribution inputs. This includes negative results where even extensive amounts of data and training time never lead to any non-trivial generalization, despite models having sufficient capacity to fit the training data perfectly. Our results show that, for our subset of tasks, RNNs and Transformers fail to generalize on non-regular tasks, LSTMs can solve regular and counter-language tasks, and only networks augmented with structured memory (such as a stack or memory tape) can successfully generalize on context-free and context-sensitive tasks.Published as a conference paper at ICLR 2023 context-sensitive recursively enumerable infinite tape finite-state controller linear tape stack counter deterministic context-free regular counter finite FFNN RNN LSTM Stack-RNN Transformer Tape-RNN Figure 1: Formal language classes and their correspondence with neural network architectures. Left: Our empirical evaluation locates the architectures on the hierarchy of formal language classes.Right: Each formal language class is associated with a minimal computational model (automaton) to recognize or generate the language (see Section 3). All automata have a finite-state controller at their core, in addition to increasingly restrictive memory access as we descend the hierarchy.practically render the model non-universal. Therefore, both architectural and training limitations impact which sequence prediction problems a model can solve in practice. In formal language theory, the Chomsky hierarchy (Chomsky, 1956) classifies such (sequence prediction) problems by increasing complexity. This hierarchy is associated with an equivalent hierarchy of models (automata) that can solve different problem classes (Savage, 1998; Sipser, 1997). Lower-level automata have restrictive memory models and can only solve lower-level problems, while Turing machines with infinite memory and unrestricted memory access lie on top of the hierachy and can solve all computable problems. However, unlike for classical automata, a unified placement of neural architectures on the Chomsky hierarchy has not yet been practically established, which is precisely the goal of our work.This workWe conduct an extensive empirical study with the aim of discovering how neural network models used for program induction relate to the idealized computational models defined by the Chomsky hierarchy in practice (seeFig. 1for a summary of our findings). We investigate whether the theoretical limitations of certain neural models hold in practice when trained with gradient-based methods. For example, previous work has theoretically argued that RNNs are Turing complete (Siegelmann & Sontag, 1994). However, more recent theoretical analyses(Ackerman & Cybenko, 2020;Merrill, 2019;Weiss et al., 2018)showed that RNNs lie much lower on the Chomsky hierarchy. To complement these theoretical analyses, we conduct a large-scale empirical evaluation on sequence prediction problems. We make the following main contributions:• We conduct an extensive generalization study (20 910 models, 15 tasks) of state-of-the-art neural network architectures (RNN, LSTM, Transformer) and memory-augmented networks (Stack-RNN, Tape-RNN) on a battery of sequence-prediction tasks spanning the entire Chomsky hierarchy that can be practically tested with finite-time computation. • We open-source a length generalization benchmark (https://github.com/deepmind/ neural _ networks _ chomsky _ hierarchy) that is out of reach for state-of-the-art sequence prediction models and allows us to pinpoint the failure modes of these architectures. • We show how increasing amounts of training data do not enable generalization on our tasks higher up in the hierarchy for some architectures (under sufficient capacity to perfectly learn the training data) potentially implying hard limitations for scaling laws (Kaplan et al., 2020). • We demonstrate how augmenting architectures with differentiable structured memory (e.g., with a stack or a tape) can enable them to solve tasks higher up the hierarchy. | Published as a conference paper at ICLR 2023 NEURAL NETWORKS AND THE CHOMSKY HIERARCHY |
d3458362 | Synthesizing realistic images from text descriptions on a dataset like Microsoft Common Objects in Context (MS COCO), where each image can contain several objects, is a challenging task. Prior work has used text captions to generate images. However, captions might not be informative enough to capture the entire image and insufficient for the model to be able to understand which objects in the images correspond to which words in the captions. We show that adding a dialogue that further describes the scene leads to significant improvement in the inception score and in the quality of generated images on the MS COCO dataset. | ChatPainter: Improving Text to Image Generation using Dialogue |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.