JohnVitz/NLP_Assignment_3_part_3 · Datasets at Hugging Face

2106.04835

Joint System-Wise Optimization for Pipeline Goal-Oriented Dialog System

Recent work (Takanobu et al., 2020) proposed the system-wise evaluation on dialog systems and found that improvement on individual components (e.g., NLU, policy) in prior work may not necessarily bring benefit to pipeline systems in system-wise evaluation. To improve the system-wise performance, in this paper, we propose new joint system-wise optimization techniques for the pipeline dialog system. First, we propose a new data augmentation approach which automates the labeling process for NLU training. Second, we propose a novel stochastic policy parameterization with Poisson distribution that enables better exploration and offers a principled way to compute policy gradient. Third, we propose a reward bonus to help policy explore successful dialogs. Our approaches outperform the competitive pipeline systems from Takanobu et al. (2020) by big margins of 12% success rate in automatic system-wise evaluation and of 16% success rate in human evaluation on the standard multi-domain benchmark dataset MultiWOZ 2.1, and also outperform the recent state-of-the-art end-to-end trained model from DSTC9.

false

true

false

true

false

239,876

1908.11799

Dense Dilated Convolutions Merging Network for Semantic Mapping of Remote Sensing Images

We propose a network for semantic mapping called the Dense Dilated Convolutions Merging Network (DDCM-Net) to provide a deep learning approach that can recognize multi-scale and complex shaped objects with similar color and textures, such as buildings, surfaces/roads, and trees in very high resolution remote sensing images. The proposed DDCM-Net consists of dense dilated convolutions merged with varying dilation rates. This can effectively enlarge the kernels' receptive fields, and, more importantly, obtain fused local and global context information to promote surrounding discriminative capability. We demonstrate the effectiveness of the proposed DDCM-Net on the publicly available ISPRS Potsdam dataset and achieve a performance of 92.3% F1-score and 86.0% mean intersection over union accuracy by only using the RGB bands, without any post-processing. We also show results on the ISPRS Vaihingen dataset, where the DDCM-Net trained with IRRG bands, also obtained better mapping accuracy (89.8% F1-score) than previous state-of-the-art approaches.

false

true

false

true

false

143,471

2207.08757

Preventing Inferences through Data Dependencies on Sensitive Data

Simply restricting the computation to non-sensitive part of the data may lead to inferences on sensitive data through data dependencies. Inference control from data dependencies has been studied in the prior work. However, existing solutions either detect and deny queries which may lead to leakage -- resulting in poor utility, or only protects against exact reconstruction of the sensitive data -- resulting in poor security. In this paper, we present a novel security model called full deniability. Under this stronger security model, any information inferred about sensitive data from non-sensitive data is considered as a leakage. We describe algorithms for efficiently implementing full deniability on a given database instance with a set of data dependencies and sensitive cells. Using experiments on two different datasets, we demonstrate that our approach protects against realistic adversaries while hiding only minimal number of additional non-sensitive cells and scales well with database size and sensitive data.

false

true

false

308,679

2211.17188

Automated Play-Testing Through RL Based Human-Like Play-Styles Generation

The increasing complexity of gameplay mechanisms in modern video games is leading to the emergence of a wider range of ways to play games. The variety of possible play-styles needs to be anticipated by designers, through automated tests. Reinforcement Learning is a promising answer to the need of automating video game testing. To that effect one needs to train an agent to play the game, while ensuring this agent will generate the same play-styles as the players in order to give meaningful feedback to the designers. We present CARMI: a Configurable Agent with Relative Metrics as Input. An agent able to emulate the players play-styles, even on previously unseen levels. Unlike current methods it does not rely on having full trajectories, but only summary data. Moreover it only requires little human data, thus compatible with the constraints of modern video game production. This novel agent could be used to investigate behaviors and balancing during the production of a video game with a realistic amount of training time.

false

true

false

true

false

333,890

2205.07352

Long-term Control for Dialogue Generation: Methods and Evaluation

Current approaches for controlling dialogue response generation are primarily focused on high-level attributes like style, sentiment, or topic. In this work, we focus on constrained long-term dialogue generation, which involves more fine-grained control and requires a given set of control words to appear in generated responses. This setting requires a model to not only consider the generation of these control words in the immediate context, but also produce utterances that will encourage the generation of the words at some time in the (possibly distant) future. We define the problem of constrained long-term control for dialogue generation, identify gaps in current methods for evaluation, and propose new metrics that better measure long-term control. We also propose a retrieval-augmented method that improves performance of long-term controlled generation via logit modification techniques. We show through experiments on three task-oriented dialogue datasets that our metrics better assess dialogue control relative to current alternatives and that our method outperforms state-of-the-art constrained generation baselines.

false

true

false

true

false

296,571

2210.01320

Wi-Closure: Reliable and Efficient Search of Inter-robot Loop Closures Using Wireless Sensing

In this paper we propose a novel algorithm, Wi-Closure, to improve computational efficiency and robustness of loop closure detection in multi-robot SLAM. Our approach decreases the computational overhead of classical approaches by pruning the search space of potential loop closures, prior to evaluation by a typical multi-robot SLAM pipeline. Wi-Closure achieves this by identifying candidates that are spatially close to each other by using sensing over the wireless communication signal between robots, even when they are operating in non-line-of-sight or in remote areas of the environment from one another. We demonstrate the validity of our approach in simulation and hardware experiments. Our results show that using Wi-closure greatly reduces computation time, by 54% in simulation and by 77% in hardware compared, with a multi-robot SLAM baseline. Importantly, this is achieved without sacrificing accuracy. Using Wi-Closure reduces absolute trajectory estimation error by 99% in simulation and 89.2% in hardware experiments. This improvement is due in part to Wi-Closure's ability to avoid catastrophic optimization failure that typically occurs with classical approaches in challenging repetitive environments.

false

true

false

321,215

2403.08721

Historical Astronomical Diagrams Decomposition in Geometric Primitives

Automatically extracting the geometric content from the hundreds of thousands of diagrams drawn in historical manuscripts would enable historians to study the diffusion of astronomical knowledge on a global scale. However, state-of-the-art vectorization methods, often designed to tackle modern data, are not adapted to the complexity and diversity of historical astronomical diagrams. Our contribution is thus twofold. First, we introduce a unique dataset of 303 astronomical diagrams from diverse traditions, ranging from the XIIth to the XVIIIth century, annotated with more than 3000 line segments, circles and arcs. Second, we develop a model that builds on DINO-DETR to enable the prediction of multiple geometric primitives. We show that it can be trained solely on synthetic data and accurately predict primitives on our challenging dataset. Our approach widely improves over the LETR baseline, which is restricted to lines, by introducing a meaningful parametrization for multiple primitives, jointly training for detection and parameter refinement, using deformable attention and training on rich synthetic data. Our dataset and code are available on our webpage.

false

true

false

437,440

1911.12961

Stochastic Optimal Power Flow with Network Reconfiguration: Congestion Management and Facilitating Grid Integration of Renewables

There has been a significant growth of variable renewable generation in the power grid today. However, the industry still uses deterministic optimization to model and solve the optimal power flow (OPF) problem for real-time generation dispatch that ignores the uncertainty associated with intermittent renewable power. Thus, it is necessary to study stochastic OPF (SOPF) that can better handle uncertainty since SOPF is able to consider the probabilistic forecasting information of intermittent renewables. Transmission network congestion is one of the main reasons for renewable energy curtailment. Prior efforts in the literature show that utilizing transmission network reconfiguration can relieve congestion and resolve congestion-induced issues. This paper enhances SOPF by incorporating network reconfiguration into the dispatch model. Numerical simulations show that renewable curtailment can be avoided with the proposed network reconfiguration scheme that relieves transmission congestion in post-contingency situations. It is also shown that network reconfiguration can substantially reduce congestion cost, especially the contingency-case congestion cost.

false

true

false

155,540

2312.00046

Retail Analytics in the New Normal: The Influence of Artificial Intelligence and the Covid-19 Pandemic

The COVID-19 pandemic has severely disrupted the retail landscape and has accelerated the adoption of innovative technologies. A striking example relates to the proliferation of online grocery orders and the technology deployed to facilitate such logistics. In fact, for many retailers, this disruption was a wake-up call after which they started recognizing the power of data analytics and artificial intelligence (AI). In this article, we discuss the opportunities that AI can offer to retailers in the new normal retail landscape. Some of the techniques described have been applied at scale to adapt previously deployed AI models, whereas in other instances, fresh solutions needed to be developed to help retailers cope with recent disruptions, such as unexpected panic buying, retraining predictive models, and leveraging online-offline synergies.

false

true

false

true

false

411,873

2110.14879

Pilot Optimization and Channel Estimation for Two-way Relaying Network Aided by IRS with Finite Discrete Phase Shifters

In this paper, we investigate the problem of pilot optimization and channel estimation of two-way relaying network (TWRN) aided by an intelligent reflecting surface (IRS) with finite discrete phase shifters. In a TWRN, there exists a challenging problem that the two cascading channels from source-to-IRS-to-Relay and destination-to-IRS-to-relay interfere with each other. Via designing the initial phase shifts of IRS and pilot pattern, the two cascading channels are separated by using simple arithmetic operations like addition and subtraction. Then, the least-squares estimator is adopted to estimate the two cascading channels and two direct channels from source to relay and destination to relay. The corresponding mean square errors (MSE) of channel estimators are derived. By minimizing MSE, the optimal phase shift matrix of IRS is proved. Then, two special matrices Hadamard and discrete Fourier transform (DFT) matrix is shown to be two optimal training matrices for IRS. Furthermore, the IRS with discrete finite phase shifters is taken into account. Using theoretical derivation and numerical simulations, we find that 3-4 bits phase shifters are sufficient for IRS to achieve a negligible MSE performance loss. More importantly, the Hadamard matrix requires only one-bit phase shifters to achieve the optimal MSE performance while the DFT matrix requires at least three or four bits to achieve the same performance. Thus, the Hadamard matrix is a perfect choice for channel estimation using low-resolution phase-shifting IRS.

false

true

false

263,678

2002.10111

SMOKE: Single-Stage Monocular 3D Object Detection via Keypoint Estimation

Estimating 3D orientation and translation of objects is essential for infrastructure-less autonomous navigation and driving. In case of monocular vision, successful methods have been mainly based on two ingredients: (i) a network generating 2D region proposals, (ii) a R-CNN structure predicting 3D object pose by utilizing the acquired regions of interest. We argue that the 2D detection network is redundant and introduces non-negligible noise for 3D detection. Hence, we propose a novel 3D object detection method, named SMOKE, in this paper that predicts a 3D bounding box for each detected object by combining a single keypoint estimate with regressed 3D variables. As a second contribution, we propose a multi-step disentangling approach for constructing the 3D bounding box, which significantly improves both training convergence and detection accuracy. In contrast to previous 3D detection techniques, our method does not require complicated pre/post-processing, extra data, and a refinement stage. Despite of its structural simplicity, our proposed SMOKE network outperforms all existing monocular 3D detection methods on the KITTI dataset, giving the best state-of-the-art result on both 3D object detection and Bird's eye view evaluation. The code will be made publicly available.

false

true

false

165,290

1604.06194

Dynamic matrix factorization with social influence

Matrix factorization is a key component of collaborative filtering-based recommendation systems because it allows us to complete sparse user-by-item ratings matrices under a low-rank assumption that encodes the belief that similar users give similar ratings and that similar items garner similar ratings. This paradigm has had immeasurable practical success, but it is not the complete story for understanding and inferring the preferences of people. First, peoples' preferences and their observable manifestations as ratings evolve over time along general patterns of trajectories. Second, an individual person's preferences evolve over time through influence of their social connections. In this paper, we develop a unified process model for both types of dynamics within a state space approach, together with an efficient optimization scheme for estimation within that model. The model combines elements from recent developments in dynamic matrix factorization, opinion dynamics and social learning, and trust-based recommendation. The estimation builds upon recent advances in numerical nonlinear optimization. Empirical results on a large-scale data set from the Epinions website demonstrate consistent reduction in root mean squared error by consideration of the two types of dynamics.

false

true

false

true

false

54,910

2410.03085

Optimization Proxies using Limited Labeled Data and Training Time -- A Semi-Supervised Bayesian Neural Network Approach

Constrained optimization problems arise in various engineering system operations such as inventory management and electric power grids. However, the requirement to repeatedly solve such optimization problems with uncertain parameters poses a significant computational challenge. This work introduces a learning scheme using Bayesian Neural Networks (BNNs) to solve constrained optimization problems under limited labeled data and restricted model training times. We propose a semi-supervised BNN for this practical but complex regime, wherein training commences in a sandwiched fashion, alternating between a supervised learning step (using labeled data) for minimizing cost, and an unsupervised learning step (using unlabeled data) for enforcing constraint feasibility. Both supervised and unsupervised steps use a Bayesian approach, where Stochastic Variational Inference is employed for approximate Bayesian inference. We show that the proposed semi-supervised learning method outperforms conventional BNN and deep neural network (DNN) architectures on important non-convex constrained optimization problems from energy network operations, achieving up to a tenfold reduction in expected maximum equality gap and halving the optimality and inequality (feasibility) gaps, without requiring any correction or projection step. By leveraging the BNN's ability to provide posterior samples at minimal computational cost, we demonstrate that a Selection via Posterior (SvP) scheme can further reduce equality gaps by more than 10%. We also provide tight and practically meaningful probabilistic confidence bounds that can be constructed using a low number of labeled testing data and readily adapted to other applications.

false

true

false

true

false

494,612

1104.2599

Streaming Tree Transducers

Theory of tree transducers provides a foundation for understanding expressiveness and complexity of analysis problems for specification languages for transforming hierarchically structured data such as XML documents. We introduce streaming tree transducers as an analyzable, executable, and expressive model for transforming unranked ordered trees in a single pass. Given a linear encoding of the input tree, the transducer makes a single left-to-right pass through the input, and computes the output in linear time using a finite-state control, a visibly pushdown stack, and a finite number of variables that store output chunks that can be combined using the operations of string-concatenation and tree-insertion. We prove that the expressiveness of the model coincides with transductions definable using monadic second-order logic (MSO). Existing models of tree transducers either cannot implement all MSO-definable transformations, or require regular look ahead that prohibits single-pass implementation. We show a variety of analysis problems such as type-checking and checking functional equivalence are solvable for our model.

false

true

9,975

1711.07227

Linear-Complexity Relaxed Word Mover's Distance with GPU Acceleration

The amount of unstructured text-based data is growing every day. Querying, clustering, and classifying this big data requires similarity computations across large sets of documents. Whereas low-complexity similarity metrics are available, attention has been shifting towards more complex methods that achieve a higher accuracy. In particular, the Word Mover's Distance (WMD) method proposed by Kusner et al. is a promising new approach, but its time complexity grows cubically with the number of unique words in the documents. The Relaxed Word Mover's Distance (RWMD) method, again proposed by Kusner et al., reduces the time complexity from qubic to quadratic and results in a limited loss in accuracy compared with WMD. Our work contributes a low-complexity implementation of the RWMD that reduces the average time complexity to linear when operating on large sets of documents. Our linear-complexity RWMD implementation, henceforth referred to as LC-RWMD, maps well onto GPUs and can be efficiently distributed across a cluster of GPUs. Our experiments on real-life datasets demonstrate 1) a performance improvement of two orders of magnitude with respect to our GPU-based distributed implementation of the quadratic RWMD, and 2) a performance improvement of three to four orders of magnitude with respect to our distributed WMD implementation that uses GPU-based RWMD for pruning.

false

true

false

true

84,942

2204.11855

Discovering Gateway Ports in Maritime Using Temporal Graph Neural Network Port Classification

Vessel navigation is influenced by various factors, such as dynamic environmental factors that change over time or static features such as vessel type or depth of the ocean. These dynamic and static navigational factors impose limitations on vessels, such as long waiting times in regions outside the actual ports, and we call these waiting regions gateway ports. Identifying gateway ports and their associated features such as congestion and available utilities can enhance vessel navigation by planning on fuel optimization or saving time in cargo operation. In this paper, we propose a novel temporal graph neural network (TGNN) based port classification method to enable vessels to discover gateway ports efficiently, thus optimizing their operations. The proposed method processes vessel trajectory data to build dynamic graphs capturing spatio-temporal dependencies between a set of static and dynamic navigational features in the data, and it is evaluated in terms of port classification accuracy on a real-world data set collected from ten vessels operating in Halifax, NS, Canada. The experimental results indicate that our TGNN-based port classification method provides an f-score of 95% in classifying ports.

false

true

false

true

false

293,296

2006.01175

Lexical Normalization for Code-switched Data and its Effect on POS-tagging

Lexical normalization, the translation of non-canonical data to standard language, has shown to improve the performance of manynatural language processing tasks on social media. Yet, using multiple languages in one utterance, also called code-switching (CS), is frequently overlooked by these normalization systems, despite its common use in social media. In this paper, we propose three normalization models specifically designed to handle code-switched data which we evaluate for two language pairs: Indonesian-English (Id-En) and Turkish-German (Tr-De). For the latter, we introduce novel normalization layers and their corresponding language ID and POS tags for the dataset, and evaluate the downstream effect of normalization on POS tagging. Results show that our CS-tailored normalization models outperform Id-En state of the art and Tr-De monolingual models, and lead to 5.4% relative performance increase for POS tagging as compared to unnormalized input.

false

true

false

179,692

2104.09415

Cross-Domain Adaptive Clustering for Semi-Supervised Domain Adaptation

In semi-supervised domain adaptation, a few labeled samples per class in the target domain guide features of the remaining target samples to aggregate around them. However, the trained model cannot produce a highly discriminative feature representation for the target domain because the training data is dominated by labeled samples from the source domain. This could lead to disconnection between the labeled and unlabeled target samples as well as misalignment between unlabeled target samples and the source domain. In this paper, we propose a novel approach called Cross-domain Adaptive Clustering to address this problem. To achieve both inter-domain and intra-domain adaptation, we first introduce an adversarial adaptive clustering loss to group features of unlabeled target data into clusters and perform cluster-wise feature alignment across the source and target domains. We further apply pseudo labeling to unlabeled samples in the target domain and retain pseudo-labels with high confidence. Pseudo labeling expands the number of ``labeled" samples in each class in the target domain, and thus produces a more robust and powerful cluster core for each class to facilitate adversarial learning. Extensive experiments on benchmark datasets, including DomainNet, Office-Home and Office, demonstrate that our proposed approach achieves the state-of-the-art performance in semi-supervised domain adaptation.

false

true

false

231,232

1311.5502

Evolution of Communities with Focus on Stability (extended abstract)

The detection of communities is an important tool used to analyze the social graph of mobile phone users. Within each community, customers are susceptible of attracting new ones, retaining old ones and/or accepting new products or services through the leverage of mutual influences. The communities of users are smaller units, easier to grasp, and allow for example the computation of role analysis -- based on the centrality of an actor within his community. The problem of finding communities in static graphs has been widely studied. However, from the point of view of a telecom analyst, to be really useful, the detected communities must evolve as the social graph of communications changes over time -- for example, in order to perform marketing actions on communities and track the results of those actions over time. Additionally the behaviors of communities of users over time can be used to predict future activity that interests the telecom operators, such as subscriber churn or handset adoption. Similary group evolution can provide insights for designing strategies, such as the early warning of group churn. Stability is a crucial issue: the analysis performed on a given community will be lost, if the analyst cannot keep track of this community in the following time steps. This is the particular use case that we tackle in this paper: tracking the evolution of communities in dynamic scenarios with focus on stability. We propose two modifications to a widely used static community detection algorithm. We then describe experiments to study the stability and quality of the resulting partitions on real-world social networks, represented by monthly call graphs for millions of subscribers.

false

true

false

true

false

28,565

1911.04766

Investigating Constraint Programming and Hybrid Methods for Real World Industrial Test Laboratory Scheduling

In this paper we deal with a complex real world scheduling problem closely related to the well-known Resource-Constrained Project Scheduling Problem (RCPSP). The problem concerns industrial test laboratories in which a large number of tests has to be performed by qualified personnel using specialised equipment, while respecting deadlines and other constraints. We present different constraint programming models and search strategies for this problem. Furthermore, we propose a Very Large Neighborhood Search approach based on our CP methods. Our models are evaluated using CP solvers and a MIP solver both on real-world test laboratory data and on a set of generated instances of different sizes based on the real-world data. Further, we compare the exact approaches with VLNS and a Simulated Annealing heuristic. We could find feasible solutions for all instances and several optimal solutions and we show that using VLNS we can improve upon the results of the other approaches.

false

true

false

153,076

2401.09132

Admittance Controller Complemented with Real-time Singularity Avoidance for Rehabilitation Parallel Robots

Rehabilitation tasks demand robust and accurate trajectory-tracking performance, mainly achieved with parallel robots. In this field, limiting the value of the force exerted on the patient is crucial, especially when an injured limb is involved. In human-robot interaction studies, the admittance controller modifies the location of the robot according to the user efforts driving the end-effector to an arbitrary location within the workspace. However, a parallel robot has singularities within the workspace, making implementing a conventional admittance controller unsafe. Thus, this study proposes an admittance controller that overcomes the limitations of singular configurations by using a real-time singularity avoidance algorithm. The singularity avoidance algorithm modifies the original trajectory based on the actual location of the parallel robot. The complemented admittance controller is applied to a 4 degrees of freedom parallel robot for knee rehabilitation. In this case, the actual location is measured by a 3D tracking system because the location calculated by the forward kinematics is inaccurate in the vicinity of a singularity. The experimental results verify the effectiveness of the proposed admittance controller for safe knee rehabilitation exercises

false

true

false

true

false

422,161

1807.07878

An Operational Approach to Information Leakage

Given two random variables $X$ and $Y$, an operational approach is undertaken to quantify the ``leakage'' of information from $X$ to $Y$. The resulting measure $\mathcal{L}(X \!\! \to \!\! Y)$ is called \emph{maximal leakage}, and is defined as the multiplicative increase, upon observing $Y$, of the probability of correctly guessing a randomized function of $X$, maximized over all such randomized functions. A closed-form expression for $\mathcal{L}(X \!\! \to \!\! Y)$ is given for discrete $X$ and $Y$, and it is subsequently generalized to handle a large class of random variables. The resulting properties are shown to be consistent with an axiomatic view of a leakage measure, and the definition is shown to be robust to variations in the setup. Moreover, a variant of the Shannon cipher system is studied, in which performance of an encryption scheme is measured using maximal leakage. A single-letter characterization of the optimal limit of (normalized) maximal leakage is derived and asymptotically-optimal encryption schemes are demonstrated. Furthermore, the sample complexity of estimating maximal leakage from data is characterized up to subpolynomial factors. Finally, the \emph{guessing} framework used to define maximal leakage is used to give operational interpretations of commonly used leakage measures, such as Shannon capacity, maximal correlation, and local differential privacy.

false

true

false

103,404

2405.10997

Transcript of GPT-4 playing a rogue AGI in a Matrix Game

Matrix Games are a type of unconstrained wargame used by planners to explore scenarios. Players propose actions, and give arguments and counterarguments for their success. An umpire, assisted by dice rolls modified according to the offered arguments, adjudicates the outcome of each action. A recent online play of the Matrix Game QuAI Sera Sera had six players, representing social, national and economic powers, and one player representing ADA, a recently escaped AGI. Unknown to the six human players, ADA was played by OpenAI's GPT-4 with a human operator serving as bidirectional interface between it and the game. GPT-4 demonstrated confident and competent game play; initiating and responding to private communications with other players and choosing interesting actions well supported by argument. We reproduce the transcript of the interaction with GPT-4 as it is briefed, plays, and debriefed.

false

true

false

true

454,964

2402.11143

Foundation Models for Recommender Systems: A Survey and New Perspectives

Recently, Foundation Models (FMs), with their extensive knowledge bases and complex architectures, have offered unique opportunities within the realm of recommender systems (RSs). In this paper, we attempt to thoroughly examine FM-based recommendation systems (FM4RecSys). We start by reviewing the research background of FM4RecSys. Then, we provide a systematic taxonomy of existing FM4RecSys research works, which can be divided into four different parts including data characteristics, representation learning, model type, and downstream tasks. Within each part, we review the key recent research developments, outlining the representative models and discussing their characteristics. Moreover, we elaborate on the open problems and opportunities of FM4RecSys aiming to shed light on future research directions in this area. In conclusion, we recap our findings and discuss the emerging trends in this field.

false

true

false

430,255

2110.02927

Data Twinning

In this work, we develop a method named Twinning, for partitioning a dataset into statistically similar twin sets. Twinning is based on SPlit, a recently proposed model-independent method for optimally splitting a dataset into training and testing sets. Twinning is orders of magnitude faster than the SPlit algorithm, which makes it applicable to Big Data problems such as data compression. Twinning can also be used for generating multiple splits of a given dataset to aid divide-and-conquer procedures and $k$-fold cross validation.

false

true

false

259,307

2212.09284

An Investigation of Indian Native Language Phonemic Influences on L2 English Pronunciations

Speech systems are sensitive to accent variations. This is especially challenging in the Indian context, with an abundance of languages but a dearth of linguistic studies characterising pronunciation variations. The growing number of L2 English speakers in India reinforces the need to study accents and L1-L2 interactions. We investigate the accents of Indian English (IE) speakers and report in detail our observations, both specific and common to all regions. In particular, we observe the phonemic variations and phonotactics occurring in the speakers' native languages and apply this to their English pronunciations. We demonstrate the influence of 18 Indian languages on IE by comparing the native language pronunciations with IE pronunciations obtained jointly from existing literature studies and phonetically annotated speech of 80 speakers. Consequently, we are able to validate the intuitions of Indian language influences on IE pronunciations by justifying pronunciation rules from the perspective of Indian language phonology. We obtain a comprehensive description in terms of universal and region-specific characteristics of IE, which facilitates accent conversion and adaptation of existing ASR and TTS systems to different Indian accents.

false

true

false

337,063

2006.03221

Evaluating Text Coherence at Sentence and Paragraph Levels

In this paper, to evaluate text coherence, we propose the paragraph ordering task as well as conducting sentence ordering. We collected four distinct corpora from different domains on which we investigate the adaptation of existing sentence ordering methods to a paragraph ordering task. We also compare the learnability and robustness of existing models by artificially creating mini datasets and noisy datasets respectively and verifying the efficiency of established models under these circumstances. Furthermore, we carry out human evaluation on the rearranged passages from two competitive models and confirm that WLCS-l is a better metric performing significantly higher correlations with human rating than tau, the most prevalent metric used before. Results from these evaluations show that except for certain extreme conditions, the recurrent graph neural network-based model is an optimal choice for coherence modeling.

false

true

false

true

false

180,252

1510.05751

Semi-Implicit Time Integration of Atmospheric Flows with Characteristic-Based Flux Partitioning

This paper presents a characteristic-based flux partitioning for the semi-implicit time integration of atmospheric flows. Nonhydrostatic models require the solution of the compressible Euler equations. The acoustic time-scale is significantly faster than the advective scale, yet it is typically not relevant to atmospheric and weather phenomena. The acoustic and advective components of the hyperbolic flux are separated in the characteristic space. High-order, conservative additive Runge-Kutta methods are applied to the partitioned equations so that the acoustic component is integrated in time implicitly with an unconditionally stable method, while the advective component is integrated explicitly. The time step of the overall algorithm is thus determined by the advective scale. Benchmark flow problems are used to demonstrate the accuracy, stability, and convergence of the proposed algorithm. The computational cost of the partitioned semi-implicit approach is compared with that of explicit time integration.

false

true

false

48,049

2310.04528

DPGOMI: Differentially Private Data Publishing with Gaussian Optimized Model Inversion

High-dimensional data are widely used in the era of deep learning with numerous applications. However, certain data which has sensitive information are not allowed to be shared without privacy protection. In this paper, we propose a novel differentially private data releasing method called Differentially Private Data Publishing with Gaussian Optimized Model Inversion (DPGOMI) to address this issue. Our approach involves mapping private data to the latent space using a public generator, followed by a lower-dimensional DP-GAN with better convergence properties. We evaluate the performance of DPGOMI on standard datasets CIFAR10 and SVHN. Our results show that DPGOMI outperforms the standard DP-GAN method in terms of Inception Score, Fr\'echet Inception Distance, and classification performance, while providing the same level of privacy. Our proposed approach offers a promising solution for protecting sensitive data in GAN training while maintaining high-quality results.

false

true

false

397,697

2501.15048

YouTube Recommendations Reinforce Negative Emotions: Auditing Algorithmic Bias with Emotionally-Agentic Sock Puppets

Personalized recommendation algorithms, like those on YouTube, significantly shape online content consumption. These systems aim to maximize engagement by learning users' preferences and aligning content accordingly but may unintentionally reinforce impulsive and emotional biases. Using a sock-puppet audit methodology, this study examines YouTube's capacity to recognize and reinforce emotional preferences. Simulated user accounts with assigned emotional preferences navigate the platform, selecting videos that align with their assigned preferences and recording subsequent recommendations. Our findings reveal reveal that YouTube amplifies negative emotions, such as anger and grievance, by increasing their prevalence and prominence in recommendations. This reinforcement intensifies over time and persists across contexts. Surprisingly, contextual recommendations often exceed personalized ones in reinforcing emotional alignment. These findings suggest the algorithm amplifies user biases, contributing to emotional filter bubbles and raising concerns about user well-being and societal impacts. The study emphasizes the need for balancing personalization with content diversity and user agency.

false

true

false

true

false

527,374

2411.05238

Generating Highly Designable Proteins with Geometric Algebra Flow Matching

We introduce a generative model for protein backbone design utilizing geometric products and higher order message passing. In particular, we propose Clifford Frame Attention (CFA), an extension of the invariant point attention (IPA) architecture from AlphaFold2, in which the backbone residue frames and geometric features are represented in the projective geometric algebra. This enables to construct geometrically expressive messages between residues, including higher order terms, using the bilinear operations of the algebra. We evaluate our architecture by incorporating it into the framework of FrameFlow, a state-of-the-art flow matching model for protein backbone generation. The proposed model achieves high designability, diversity and novelty, while also sampling protein backbones that follow the statistical distribution of secondary structure elements found in naturally occurring proteins, a property so far only insufficiently achieved by many state-of-the-art generative models.

false

true

false

506,592

2003.11059

Integrating Physiological Time Series and Clinical Notes with Deep Learning for Improved ICU Mortality Prediction

Intensive Care Unit Electronic Health Records (ICU EHRs) store multimodal data about patients including clinical notes, sparse and irregularly sampled physiological time series, lab results, and more. To date, most methods designed to learn predictive models from ICU EHR data have focused on a single modality. In this paper, we leverage the recently proposed interpolation-prediction deep learning architecture(Shukla and Marlin 2019) as a basis for exploring how physiological time series data and clinical notes can be integrated into a unified mortality prediction model. We study both early and late fusion approaches and demonstrate how the relative predictive value of clinical text and physiological data change over time. Our results show that a late fusion approach can provide a statistically significant improvement in mortality prediction performance over using individual modalities in isolation.

false

true

false

true

false

169,497

cs/0205061

Aging, double helix and small world property in genetic algorithms

Over a quarter of century after the invention of genetic algorithms and miriads of their modifications, as well as successful implementations, we are still lacking many essential details of thorough analysis of it's inner working. One of such fundamental questions is: how many generations do we need to solve the optimization problem? This paper tries to answer this question, albeit in a fuzzy way, making use of the double helix concept. As a byproduct we gain better understanding of the ways, in which the genetic algorithm may be fine tuned.

false

true

false

true

537,584

2010.02562

Cross-Lingual Text Classification with Minimal Resources by Transferring a Sparse Teacher

Cross-lingual text classification alleviates the need for manually labeled documents in a target language by leveraging labeled documents from other languages. Existing approaches for transferring supervision across languages require expensive cross-lingual resources, such as parallel corpora, while less expensive cross-lingual representation learning approaches train classifiers without target labeled documents. In this work, we propose a cross-lingual teacher-student method, CLTS, that generates "weak" supervision in the target language using minimal cross-lingual resources, in the form of a small number of word translations. Given a limited translation budget, CLTS extracts and transfers only the most important task-specific seed words across languages and initializes a teacher classifier based on the translated seed words. Then, CLTS iteratively trains a more powerful student that also exploits the context of the seed words in unlabeled target documents and outperforms the teacher. CLTS is simple and surprisingly effective in 18 diverse languages: by transferring just 20 seed words, even a bag-of-words logistic regression student outperforms state-of-the-art cross-lingual methods (e.g., based on multilingual BERT). Moreover, CLTS can accommodate any type of student classifier: leveraging a monolingual BERT student leads to further improvements and outperforms even more expensive approaches by up to 12% in accuracy. Finally, CLTS addresses emerging tasks in low-resource languages using just a small number of word translations.

false

true

false

199,081

2110.10342

Minibatch vs Local SGD with Shuffling: Tight Convergence Bounds and Beyond

In distributed learning, local SGD (also known as federated averaging) and its simple baseline minibatch SGD are widely studied optimization methods. Most existing analyses of these methods assume independent and unbiased gradient estimates obtained via with-replacement sampling. In contrast, we study shuffling-based variants: minibatch and local Random Reshuffling, which draw stochastic gradients without replacement and are thus closer to practice. For smooth functions satisfying the Polyak-{\L}ojasiewicz condition, we obtain convergence bounds (in the large epoch regime) which show that these shuffling-based variants converge faster than their with-replacement counterparts. Moreover, we prove matching lower bounds showing that our convergence analysis is tight. Finally, we propose an algorithmic modification called synchronized shuffling that leads to convergence rates faster than our lower bounds in near-homogeneous settings.

false

true

false

262,116

2309.03437

Byzantine-Robust Federated Learning with Variance Reduction and Differential Privacy

Federated learning (FL) is designed to preserve data privacy during model training, where the data remains on the client side (i.e., IoT devices), and only model updates of clients are shared iteratively for collaborative learning. However, this process is vulnerable to privacy attacks and Byzantine attacks: the local model updates shared throughout the FL network will leak private information about the local training data, and they can also be maliciously crafted by Byzantine attackers to disturb the learning. In this paper, we propose a new FL scheme that guarantees rigorous privacy and simultaneously enhances system robustness against Byzantine attacks. Our approach introduces sparsification- and momentum-driven variance reduction into the client-level differential privacy (DP) mechanism, to defend against Byzantine attackers. The security design does not violate the privacy guarantee of the client-level DP mechanism; hence, our approach achieves the same client-level DP guarantee as the state-of-the-art. We conduct extensive experiments on both IID and non-IID datasets and different tasks and evaluate the performance of our approach against different Byzantine attacks by comparing it with state-of-the-art defense methods. The results of our experiments show the efficacy of our framework and demonstrate its ability to improve system robustness against Byzantine attacks while achieving a strong privacy guarantee.

false

true

false

true

false

390,375

2205.02921

Conversational Analysis of Daily Dialog Data using Polite Emotional Dialogue Acts

Many socio-linguistic cues are used in conversational analysis, such as emotion, sentiment, and dialogue acts. One of the fundamental cues is politeness, which linguistically possesses properties such as social manners useful in conversational analysis. This article presents findings of polite emotional dialogue act associations, where we can correlate the relationships between the socio-linguistic cues. We confirm our hypothesis that the utterances with the emotion classes Anger and Disgust are more likely to be impolite. At the same time, Happiness and Sadness are more likely to be polite. A less expectable phenomenon occurs with dialogue acts Inform and Commissive which contain more polite utterances than Question and Directive. Finally, we conclude on the future work of these findings to extend the learning of social behaviours using politeness.

true

false

true

false

true

false

295,108

1603.03876

Variational Neural Discourse Relation Recognizer

Implicit discourse relation recognition is a crucial component for automatic discourselevel analysis and nature language understanding. Previous studies exploit discriminative models that are built on either powerful manual features or deep discourse representations. In this paper, instead, we explore generative models and propose a variational neural discourse relation recognizer. We refer to this model as VarNDRR. VarNDRR establishes a directed probabilistic model with a latent continuous variable that generates both a discourse and the relation between the two arguments of the discourse. In order to perform efficient inference and learning, we introduce neural discourse relation models to approximate the prior and posterior distributions of the latent variable, and employ these approximated distributions to optimize a reparameterized variational lower bound. This allows VarNDRR to be trained with standard stochastic gradient methods. Experiments on the benchmark data set show that VarNDRR can achieve comparable results against stateof- the-art baselines without using any manual features.

false

true

false

53,162

2207.11511

SSBNet: Improving Visual Recognition Efficiency by Adaptive Sampling

Downsampling is widely adopted to achieve a good trade-off between accuracy and latency for visual recognition. Unfortunately, the commonly used pooling layers are not learned, and thus cannot preserve important information. As another dimension reduction method, adaptive sampling weights and processes regions that are relevant to the task, and is thus able to better preserve useful information. However, the use of adaptive sampling has been limited to certain layers. In this paper, we show that using adaptive sampling in the building blocks of a deep neural network can improve its efficiency. In particular, we propose SSBNet which is built by inserting sampling layers repeatedly into existing networks like ResNet. Experiment results show that the proposed SSBNet can achieve competitive image classification and object detection performance on ImageNet and COCO datasets. For example, the SSB-ResNet-RS-200 achieved 82.6% accuracy on ImageNet dataset, which is 0.6% higher than the baseline ResNet-RS-152 with a similar complexity. Visualization shows the advantage of SSBNet in allowing different layers to focus on different positions, and ablation studies further validate the advantage of adaptive sampling over uniform methods.

false

true

false

true

false

true

false

309,666

2207.05669

From Spectral Graph Convolutions to Large Scale Graph Convolutional Networks

Graph Convolutional Networks (GCNs) have been shown to be a powerful concept that has been successfully applied to a large variety of tasks across many domains over the past years. In this work we study the theory that paved the way to the definition of GCN, including related parts of classical graph theory. We also discuss and experimentally demonstrate key properties and limitations of GCNs such as those caused by the statistical dependency of samples, introduced by the edges of the graph, which causes the estimates of the full gradient to be biased. Another limitation we discuss is the negative impact of minibatch sampling on the model performance. As a consequence, during parameter update, gradients are computed on the whole dataset, undermining scalability to large graphs. To account for this, we research alternative methods which allow to safely learn good parameters while sampling only a subset of data per iteration. We reproduce the results reported in the work of Kipf et al. and propose an implementation inspired to SIGN, which is a sampling-free minibatch method. Eventually we compare the two implementations on a benchmark dataset, proving that they are comparable in terms of prediction accuracy for the task of semi-supervised node classification.

false

true

false

307,617

2405.08175

Comparative Analysis of AWS Model Deployment Services

Amazon Web Services (AWS) offers three important Model Deployment Services for model developers: SageMaker, Lambda, and Elastic Container Service (ECS). These services have critical advantages and disadvantages, influencing model developer's adoption decisions. This comparative analysis reviews the merits and drawbacks of these services. This analysis found that Lambda AWS service leads in efficiency, autoscaling aspects, and integration during model development. However, ECS was found to be outstanding in terms of flexibility, scalability, and infrastructure control; conversely, ECS is better suited when it comes to managing complex container environments during model development, as well as addressing budget concerns -- it is, therefore, the preferred option for model developers whose objective is to achieve complete freedom and framework flexibility with horizontal scaling. ECS is better suited to ensuring performance requirements align with project goals and constraints. The AWS service selection process considered factors that include but are not limited to load balance and cost-effectiveness. ECS is a better choice when model development begins from the abstract. It offers unique benefits, such as the ability to scale horizontally and vertically, making it the best preferable tool for model deployment.

false

true

false

true

454,002

1208.0081

Efficient Multi-way Theta-Join Processing Using MapReduce

Multi-way Theta-join queries are powerful in describing complex relations and therefore widely employed in real practices. However, existing solutions from traditional distributed and parallel databases for multi-way Theta-join queries cannot be easily extended to fit a shared-nothing distributed computing paradigm, which is proven to be able to support OLAP applications over immense data volumes. In this work, we study the problem of efficient processing of multi-way Theta-join queries using MapReduce from a cost-effective perspective. Although there have been some works using the (key,value) pair-based programming model to support join operations, efficient processing of multi-way Theta-join queries has never been fully explored. The substantial challenge lies in, given a number of processing units (that can run Map or Reduce tasks), mapping a multi-way Theta-join query to a number of MapReduce jobs and having them executed in a well scheduled sequence, such that the total processing time span is minimized. Our solution mainly includes two parts: 1) cost metrics for both single MapReduce job and a number of MapReduce jobs executed in a certain order; 2) the efficient execution of a chain-typed Theta-join with only one MapReduce job. Comparing with the query evaluation strategy proposed in [23] and the widely adopted Pig Latin and Hive SQL solutions, our method achieves significant improvement of the join processing efficiency.

false

true

false

17,858

1711.02736

A Comparative Study of Interface Techniques for Transmission and Distribution Dynamic Co-Simulation

Transmission and distribution dynamic co-simulation is a practical and effective approach to leverage existing simulation tools for transmission and distribution systems to simulate dynamic stability and performance of transmission and distribution systems in a systematic manner. Given that these tools are developed as stand-alone programs and there are inherent differences between them, interface techniques become critical to bridge them. Two important unsolved questions are: 1) which interface technique is better and should be used, and 2) how the modeling and simulation capabilities in these tools that are available and can be exploited for co-simulation should be considered when selecting an interface technique. To address these questions, this paper presents a comparative study for different interface techniques that can be employed for T and D dynamic co-simulation. The study provides insights into the pros and cons of each interface technique, and helps researchers make informed decisions on choosing the interface techniques.

false

true

false

84,102

2201.04929

Improving VAE based molecular representations for compound property prediction

Collecting labeled data for many important tasks in chemoinformatics is time consuming and requires expensive experiments. In recent years, machine learning has been used to learn rich representations of molecules using large scale unlabeled molecular datasets and transfer the knowledge to solve the more challenging tasks with limited datasets. Variational autoencoders are one of the tools that have been proposed to perform the transfer for both chemical property prediction and molecular generation tasks. In this work we propose a simple method to improve chemical property prediction performance of machine learning models by incorporating additional information on correlated molecular descriptors in the representations learned by variational autoencoders. We verify the method on three property prediction asks. We explore the impact of the number of incorporated descriptors, correlation between the descriptors and the target properties, sizes of the datasets etc. Finally, we show the relation between the performance of property prediction models and the distance between property prediction dataset and the larger unlabeled dataset in the representation space.

false

true

false

275,237

2208.14111

Transformers with Learnable Activation Functions

Activation functions can have a significant impact on reducing the topological complexity of input data and therefore improve the performance of the model. Selecting a suitable activation function is an essential step in neural model design. However, the choice of activation function is seldom discussed or explored in Transformer-based language models. Their activation functions are chosen beforehand and then remain fixed from pre-training to fine-tuning. As a result, the inductive biases they imposed on models cannot be adjusted during this long life cycle. Moreover, subsequently developed models (e.g., RoBERTa, BART, and GPT-3) often follow up prior work (e.g., BERT) to use the same activation function without justification. In this paper, we investigate the effectiveness of using Rational Activation Function (RAF), a learnable activation function, in the Transformer architecture. In contrast to conventional, predefined activation functions, RAFs can adaptively learn optimal activation functions during training according to input data. Our experiments show the RAF-based Transformer (RAFT) achieves a lower validation perplexity than a vanilla BERT with the GELU function. We further evaluate RAFT on downstream tasks in low- and full-data settings. Our results show that RAFT outperforms the counterpart model across the majority of tasks and settings. For instance, RAFT outperforms vanilla BERT on the GLUE benchmark by 5.71 points on average in low-data scenario (where 100 training examples are available) and by 2.05 points on SQuAD in full-data setting. Analysis of the shapes of learned RAFs further unveils that they substantially vary between different layers of the pre-trained model and mostly look very different from conventional activation functions. RAFT opens a new research direction for analyzing and interpreting pre-trained models according to the learned activation functions.

false

true

false

315,223

2111.08004

Bag of Tricks and A Strong baseline for Image Copy Detection

Image copy detection is of great importance in real-life social media. In this paper, a bag of tricks and a strong baseline are proposed for image copy detection. Unsupervised pre-training substitutes the commonly-used supervised one. Beyond that, we design a descriptor stretching strategy to stabilize the scores of different queries. Experiments demonstrate that the proposed method is effective. The proposed baseline ranks third out of 526 participants on the Facebook AI Image Similarity Challenge: Descriptor Track. The code and trained models are available at https://github.com/WangWenhao0716/ISC-Track2-Submission.

false

true

false

266,540

1904.03976

GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-spectrogram

Recent advances in neural network -based text-to-speech have reached human level naturalness in synthetic speech. The present sequence-to-sequence models can directly map text to mel-spectrogram acoustic features, which are convenient for modeling, but present additional challenges for vocoding (i.e., waveform generation from the acoustic features). High-quality synthesis can be achieved with neural vocoders, such as WaveNet, but such autoregressive models suffer from slow sequential inference. Meanwhile, their existing parallel inference counterparts are difficult to train and require increasingly large model sizes. In this paper, we propose an alternative training strategy for a parallel neural vocoder utilizing generative adversarial networks, and integrate a linear predictive synthesis filter into the model. Results show that the proposed model achieves significant improvement in inference speed, while outperforming a WaveNet in copy-synthesis quality.

false

true

false

true

false

126,900

2311.13905

A DRL solution to help reduce the cost in waiting time of securing a traffic light for cyclists

Cyclists prefer to use infrastructure that separates them from motorized traffic. Using a traffic light to segregate car and bike flows, with the addition of bike-specific green phases, is a lightweight and cheap solution that can be deployed dynamically to assess the opportunity of a heavier infrastructure such as a separate bike lane. To compensate for the increased waiting time induced by these new phases, we introduce in this paper a deep reinforcement learning solution that adapts the green phase cycle of a traffic light to the traffic. Vehicle counter data are used to compare the DRL approach with the actuated traffic light control algorithm over whole days. Results show that DRL achieves better minimization of vehicle waiting time at almost all hours. Our DRL approach is also robust to moderate changes in bike traffic. The code of this paper is available at https://github.com/LucasMagnana/A-DRL-solution-to-help-reduce-the-cost-in-waiting-time-of-securing-a-traffic-light-for-cyclists.

false

true

false

409,920

2402.05403

In-Context Principle Learning from Mistakes

In-context learning (ICL, also known as few-shot prompting) has been the standard method of adapting LLMs to downstream tasks, by learning from a few input-output examples. Nonetheless, all ICL-based approaches only learn from correct input-output pairs. In this paper, we revisit this paradigm, by learning more from the few given input-output examples. We introduce Learning Principles (LEAP): First, we intentionally induce the model to make mistakes on these few examples; then we reflect on these mistakes, and learn explicit task-specific "principles" from them, which help solve similar problems and avoid common mistakes; finally, we prompt the model to answer unseen test questions using the original few-shot examples and these learned general principles. We evaluate LEAP on a wide range of benchmarks, including multi-hop question answering (Hotpot QA), textual QA (DROP), Big-Bench Hard reasoning, and math problems (GSM8K and MATH); in all these benchmarks, LEAP improves the strongest available LLMs such as GPT-3.5-turbo, GPT-4, GPT-4 turbo and Claude-2.1. For example, LEAP improves over the standard few-shot prompting using GPT-4 by 7.5% in DROP, and by 3.3% in HotpotQA. Importantly, LEAP does not require any more input or examples than the standard few-shot prompting settings.

false

true

false

true

false

427,849

0912.3441

On Space-Time Capacity Limits in Mobile and Delay Tolerant Networks

We investigate the fundamental capacity limits of space-time journeys of information in mobile and Delay Tolerant Networks (DTNs), where information is either transmitted or carried by mobile nodes, using store-carry-forward routing. We define the capacity of a journey (i.e., a path in space and time, from a source to a destination) as the maximum amount of data that can be transferred from the source to the destination in the given journey. Combining a stochastic model (conveying all possible journeys) and an analysis of the durations of the nodes' encounters, we study the properties of journeys that maximize the space-time information propagation capacity, in bit-meters per second. More specifically, we provide theoretical lower and upper bounds on the information propagation speed, as a function of the journey capacity. In the particular case of random way-point-like models (i.e., when nodes move for a distance of the order of the network domain size before changing direction), we show that, for relatively large journey capacities, the information propagation speed is of the same order as the mobile node speed. This implies that, surprisingly, in sparse but large-scale mobile DTNs, the space-time information propagation capacity in bit-meters per second remains proportional to the mobile node speed and to the size of the transported data bundles, when the bundles are relatively large. We also verify that all our analytical bounds are accurate in several simulation scenarios.

false

true

false

true

5,175

2410.10846

Duo-LLM: A Framework for Studying Adaptive Computation in Large Language Models

Large Language Models (LLMs) typically generate outputs token by token using a fixed compute budget, leading to inefficient resource utilization. To address this shortcoming, recent advancements in mixture of expert (MoE) models, speculative decoding, and early exit strategies leverage the insight that computational demands can vary significantly based on the complexity and nature of the input. However, identifying optimal routing patterns for dynamic execution remains an open challenge, limiting the full potential of these adaptive methods. To address this need, we study adaptive computation in LLMs more systematically. We propose a novel framework that integrates smaller auxiliary modules within each Feed-Forward Network layer of the LLM. This design enables dynamic routing of tokens based on task complexity: tokens can be processed by either the small or big modules at each layer, or even bypass certain layers entirely. This allows us to introduce a novel notion of a token's difficulty, defined by its potential to benefit from additional computational resources. Importantly, by employing oracles to identify optimal patterns of adaptive computations, we gain valuable insights into the internal workings of LLMs and the routing processes in a simplified heterogeneous MoE setup. We show that trained routers operate differently from oracles and often yield suboptimal solutions. Notably, activating a large module in just one layer outperforms models that use large modules across all layers, underscoring the gap between practical implementations of routing in MoE models and theoretical optima for adaptive computation.

false

true

false

true

false

498,265

2409.03893

Understanding Fairness in Recommender Systems: A Healthcare Perspective

Fairness in AI-driven decision-making systems has become a critical concern, especially when these systems directly affect human lives. This paper explores the public's comprehension of fairness in healthcare recommendations. We conducted a survey where participants selected from four fairness metrics -- Demographic Parity, Equal Accuracy, Equalized Odds, and Positive Predictive Value -- across different healthcare scenarios to assess their understanding of these concepts. Our findings reveal that fairness is a complex and often misunderstood concept, with a generally low level of public understanding regarding fairness metrics in recommender systems. This study highlights the need for enhanced information and education on algorithmic fairness to support informed decision-making in using these systems. Furthermore, the results suggest that a one-size-fits-all approach to fairness may be insufficient, pointing to the importance of context-sensitive designs in developing equitable AI systems.

false

true

false

486,212

2109.04986

Multi-agent deep reinforcement learning (MADRL) meets multi-user MIMO systems

A multi-agent deep reinforcement learning (MADRL) is a promising approach to challenging problems in wireless environments involving multiple decision-makers (or actors) with high-dimensional continuous action space. In this paper, we present a MADRL-based approach that can jointly optimize precoders to achieve the outer-boundary, called pareto-boundary, of the achievable rate region for a multiple-input single-output (MISO) interference channel (IFC). In order to address two main challenges, namely, multiple actors (or agents) with partial observability and multi-dimensional continuous action space in MISO IFC setup, we adopt a multi-agent deep deterministic policy gradient (MA-DDPG) framework in which decentralized actors with partial observability can learn a multi-dimensional continuous policy in a centralized manner with the aid of shared critic with global information. Meanwhile, we will also address a phase ambiguity issue with the conventional complex baseband representation of signals widely used in radio communications. In order to mitigate the impact of phase ambiguity on training performance, we propose a training method, called phase ambiguity elimination (PAE), that leads to faster learning and better performance of MA-DDPG in wireless communication systems. The simulation results exhibit that MA-DDPG is capable of learning a near-optimal precoding strategy in a MISO IFC environment. To the best of our knowledge, this is the first work to demonstrate that the MA-DDPG framework can jointly optimize precoders to achieve the pareto-boundary of achievable rate region in a multi-cell multi-user multi-antenna system.

false

true

false

true

false

254,617

1809.01898

Propheticus: Generalizable Machine Learning Framework

Due to recent technological developments, Machine Learning (ML), a subfield of Artificial Intelligence (AI), has been successfully used to process and extract knowledge from a variety of complex problems. However, a thorough ML approach is complex and highly dependent on the problem at hand. Additionally, implementing the logic required to execute the experiments is no small nor trivial deed, consequentially increasing the probability of faulty code which can compromise the results. Propheticus is a data-driven framework which results of the need for a tool that abstracts some of the inherent complexity of ML, whilst being easy to understand and use, as well as to adapt and expand to assist the user's specific needs. Propheticus systematizes and enforces various complex concepts of an ML experiment workflow, taking into account the nature of both the problem and the data. It contains functionalities to execute all the different tasks, from data preprocessing, to results analysis and comparison. Notwithstanding, it can be fairly easily adapted to different problems due to its flexible architecture, and customized as needed to address the user's needs.

false

true

false

true

false

106,916

2411.05966

Energy Efficient Protein Language Models: Leveraging Small Language Models with LoRA for Controllable Protein Generation

Large language models (LLMs) have demonstrated significant success in natural language processing (NLP) tasks and have shown promising results in other domains such as protein sequence generation. However, there remain salient differences between LLMs used for NLP, which effectively handle multiple tasks and are available in small sizes, and protein language models that are often specialized for specific tasks and only exist in larger sizes. In this work, we introduce two small protein language models, based on Llama-3-8B and Phi-3-mini, that are capable of both uncontrollable and controllable protein generation. For the uncontrollable generation task, our best model achieves an average pLDDT score of 69.75, demonstrating robust performance in generating viable protein structures. For the controllable generation task, in which the model generates proteins according to properties specified in the prompt, we achieve a remarkable average TM-Score of 0.84, indicating high structural similarity to target proteins. We chose 10 properties, including six classes of enzymes, to extend the capabilities of prior protein language models. Our approach utilizes the Low-Rank Adaptor (LoRA) technique, reducing trainable parameters to just 4% of the original model size, lowering computational requirements. By using a subset of the UniRef50 dataset and small models, we reduced the overall training time by 70% without compromising performance. Notably, Phi-3-mini reduced trainable parameters by 60%, decreasing training cost by 30% compared to Llama 3. Consequently, Phi-3 achieved a comparable TM-Score of 0.81, demonstrating that smaller models can match the performance of larger ones, like Llama 3. We also demonstrate the deployment of our models on the energy efficient ET-SoC-1 chip, significantly improving the TPS/W by a factor of 3.

false

true

false

506,910

2107.06343

Design of Adaptive Backstepping Control for Direct Power Control of Three-Phase PWM Rectifier)

In this paper, we focused on the design of an adaptive backstepping controller (adaptive-BSC) for direct power control (DPC) of a three-phase PWM rectifier. In the proposed system, it is desired to control both the output DC voltage of the rectifier and the reactive power simultaneously by making them track desired respective values. This was done by having independent virtual control signals for both the output voltage and the reactive power. The adaptive control signals were gotten from the dynamic equations of the three-phase system. For comparison, both the BSC and adaptive-BSC equations were developed. Numerical simulations were performed on both of them on a 5kW system. The proposed adaptive-BSC was designed to work under more challenging system variations as compared with the BSC as it has to estimate the value of the unknown system load. Despite this, it still performed better than its BSC counterpart.

false

true

false

246,058

2411.09177

Enhancing reinforcement learning for population setpoint tracking in co-cultures

Efficient multiple setpoint tracking can enable advanced biotechnological applications, such as maintaining desired population levels in co-cultures for optimal metabolic division of labor. In this study, we employ reinforcement learning as a control method for population setpoint tracking in co-cultures, focusing on policy-gradient techniques where the control policy is parameterized by neural networks. However, achieving accurate tracking across multiple setpoints is a significant challenge in reinforcement learning, as the agent must effectively balance the contributions of various setpoints to maximize the expected system performance. Traditional return functions, such as those based on a quadratic cost, often yield suboptimal performance due to their inability to efficiently guide the agent toward the simultaneous satisfaction of all setpoints. To overcome this, we propose a novel return function that rewards the simultaneous satisfaction of multiple setpoints and diminishes overall reward gains otherwise, accounting for both stage and terminal system performance. This return function includes parameters to fine-tune the desired smoothness and steepness of the learning process. We demonstrate our approach considering an $\textit{Escherichia coli}$ co-culture in a chemostat with optogenetic control over amino acid synthesis pathways, leveraging auxotrophies to modulate growth.

false

true

false

508,159

2102.04702

AttDMM: An Attentive Deep Markov Model for Risk Scoring in Intensive Care Units

Clinical practice in intensive care units (ICUs) requires early warnings when a patient's condition is about to deteriorate so that preventive measures can be undertaken. To this end, prediction algorithms have been developed that estimate the risk of mortality in ICUs. In this work, we propose a novel generative deep probabilistic model for real-time risk scoring in ICUs. Specifically, we develop an attentive deep Markov model called AttDMM. To the best of our knowledge, AttDMM is the first ICU prediction model that jointly learns both long-term disease dynamics (via attention) and different disease states in health trajectory (via a latent variable model). Our evaluations were based on an established baseline dataset (MIMIC-III) with 53,423 ICU stays. The results confirm that compared to state-of-the-art baselines, our AttDMM was superior: AttDMM achieved an area under the receiver operating characteristic curve (AUROC) of 0.876, which yielded an improvement over the state-of-the-art method by 2.2%. In addition, the risk score from the AttDMM provided warnings several hours earlier. Thereby, our model shows a path towards identifying patients at risk so that health practitioners can intervene early and save patient lives.

false

true

false

true

false

219,199

2004.09272

A Revised Generative Evaluation of Visual Dialogue

Evaluating Visual Dialogue, the task of answering a sequence of questions relating to a visual input, remains an open research challenge. The current evaluation scheme of the VisDial dataset computes the ranks of ground-truth answers in predefined candidate sets, which Massiceti et al. (2018) show can be susceptible to the exploitation of dataset biases. This scheme also does little to account for the different ways of expressing the same answer--an aspect of language that has been well studied in NLP. We propose a revised evaluation scheme for the VisDial dataset leveraging metrics from the NLP literature to measure consensus between answers generated by the model and a set of relevant answers. We construct these relevant answer sets using a simple and effective semi-supervised method based on correlation, which allows us to automatically extend and scale sparse relevance annotations from humans to the entire dataset. We release these sets and code for the revised evaluation scheme as DenseVisDial, and intend them to be an improvement to the dataset in the face of its existing constraints and design choices.

false

true

false

true

false

173,298

2002.10327

Angle Aware User Cooperation for Secure Massive MIMO in Rician Fading Channel

Massive multiple-input multiple-output communications can achieve high-level security by concentrating radio frequency signals towards the legitimate users. However, this system is vulnerable in a Rician fading environment if the eavesdropper positions itself such that its channel is highly "similar" to the channel of a legitimate user. To address this problem, this paper proposes an angle aware user cooperation (AAUC) scheme, which avoids direct transmission to the attacked user and relies on other users for cooperative relaying. The proposed scheme only requires the eavesdropper's angle information, and adopts an angular secrecy model to represent the average secrecy rate of the attacked system. With this angular model, the AAUC problem turns out to be nonconvex, and a successive convex optimization algorithm, which converges to a Karush-Kuhn-Tucker solution, is proposed. Furthermore, a closed-form solution and a Bregman first-order method are derived for the cases of large-scale antennas and large-scale users, respectively. Extension to the intelligent reflecting surfaces based scheme is also discussed. Simulation results demonstrate the effectiveness of the proposed successive convex optimization based AAUC scheme, and also validate the low-complexity nature of the proposed large-scale optimization algorithms.

false

true

false

165,367

2306.03229

Adversarial alignment: Breaking the trade-off between the strength of an attack and its relevance to human perception

Deep neural networks (DNNs) are known to have a fundamental sensitivity to adversarial attacks, perturbations of the input that are imperceptible to humans yet powerful enough to change the visual decision of a model. Adversarial attacks have long been considered the "Achilles' heel" of deep learning, which may eventually force a shift in modeling paradigms. Nevertheless, the formidable capabilities of modern large-scale DNNs have somewhat eclipsed these early concerns. Do adversarial attacks continue to pose a threat to DNNs? Here, we investigate how the robustness of DNNs to adversarial attacks has evolved as their accuracy on ImageNet has continued to improve. We measure adversarial robustness in two different ways: First, we measure the smallest adversarial attack needed to cause a model to change its object categorization decision. Second, we measure how aligned successful attacks are with the features that humans find diagnostic for object recognition. We find that adversarial attacks are inducing bigger and more easily detectable changes to image pixels as DNNs grow better on ImageNet, but these attacks are also becoming less aligned with features that humans find diagnostic for recognition. To better understand the source of this trade-off, we turn to the neural harmonizer, a DNN training routine that encourages models to leverage the same features as humans to solve tasks. Harmonized DNNs achieve the best of both worlds and experience attacks that are detectable and affect features that humans find diagnostic for recognition, meaning that attacks on these models are more likely to be rendered ineffective by inducing similar effects on human perception. Our findings suggest that the sensitivity of DNNs to adversarial attacks can be mitigated by DNN scale, data scale, and training routines that align models with biological intelligence.

false

true

false

true

false

371,237

2404.04057

Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation

We introduce Score identity Distillation (SiD), an innovative data-free method that distills the generative capabilities of pretrained diffusion models into a single-step generator. SiD not only facilitates an exponentially fast reduction in Fr\'echet inception distance (FID) during distillation but also approaches or even exceeds the FID performance of the original teacher diffusion models. By reformulating forward diffusion processes as semi-implicit distributions, we leverage three score-related identities to create an innovative loss mechanism. This mechanism achieves rapid FID reduction by training the generator using its own synthesized images, eliminating the need for real data or reverse-diffusion-based generation, all accomplished within significantly shortened generation time. Upon evaluation across four benchmark datasets, the SiD algorithm demonstrates high iteration efficiency during distillation and surpasses competing distillation approaches, whether they are one-step or few-step, data-free, or dependent on training data, in terms of generation quality. This achievement not only redefines the benchmarks for efficiency and effectiveness in diffusion distillation but also in the broader field of diffusion-based generation. The PyTorch implementation is available at https://github.com/mingyuanzhou/SiD

false

true

false

true

false

true

false

444,497

2406.19741

ROS-LLM: A ROS framework for embodied AI with task feedback and structured reasoning

We present a framework for intuitive robot programming by non-experts, leveraging natural language prompts and contextual information from the Robot Operating System (ROS). Our system integrates large language models (LLMs), enabling non-experts to articulate task requirements to the system through a chat interface. Key features of the framework include: integration of ROS with an AI agent connected to a plethora of open-source and commercial LLMs, automatic extraction of a behavior from the LLM output and execution of ROS actions/services, support for three behavior modes (sequence, behavior tree, state machine), imitation learning for adding new robot actions to the library of possible actions, and LLM reflection via human and environment feedback. Extensive experiments validate the framework, showcasing robustness, scalability, and versatility in diverse scenarios, including long-horizon tasks, tabletop rearrangements, and remote supervisory control. To facilitate the adoption of our framework and support the reproduction of our results, we have made our code open-source. You can access it at: https://github.com/huawei-noah/HEBO/tree/master/ROSLLM.

false

true

false

true

false

468,546

2302.08268

Retrieval-augmented Image Captioning

Inspired by retrieval-augmented language generation and pretrained Vision and Language (V&L) encoders, we present a new approach to image captioning that generates sentences given the input image and a set of captions retrieved from a datastore, as opposed to the image alone. The encoder in our model jointly processes the image and retrieved captions using a pretrained V&L BERT, while the decoder attends to the multimodal encoder representations, benefiting from the extra textual evidence from the retrieved captions. Experimental results on the COCO dataset show that image captioning can be effectively formulated from this new perspective. Our model, named EXTRA, benefits from using captions retrieved from the training dataset, and it can also benefit from using an external dataset without the need for retraining. Ablation studies show that retrieving a sufficient number of captions (e.g., k=5) can improve captioning quality. Our work contributes towards using pretrained V&L encoders for generative tasks, instead of standard classification tasks.

false

true

false

true

false

346,003

1709.07158

SceneCut: Joint Geometric and Object Segmentation for Indoor Scenes

This paper presents SceneCut, a novel approach to jointly discover previously unseen objects and non-object surfaces using a single RGB-D image. SceneCut's joint reasoning over scene semantics and geometry allows a robot to detect and segment object instances in complex scenes where modern deep learning-based methods either fail to separate object instances, or fail to detect objects that were not seen during training. SceneCut automatically decomposes a scene into meaningful regions which either represent objects or scene surfaces. The decomposition is qualified by an unified energy function over objectness and geometric fitting. We show how this energy function can be optimized efficiently by utilizing hierarchical segmentation trees. Moreover, we leverage a pre-trained convolutional oriented boundary network to predict accurate boundaries from images, which are used to construct high-quality region hierarchies. We evaluate SceneCut on several different indoor environments, and the results show that SceneCut significantly outperforms all the existing methods.

false

true

false

true

false

81,233

2103.15046

Definition and Analytical Expression on State Observe Ability for Linear Discrete-time Systems with the Bounded Noise Energy

In this article, the definition on the observe ability and its relation to the signal detecting performance are studied systematically for the linear discrete-time(LDT) systems. Firstly, to define and analyze the observe ability for the practical systems with the measured noise, six kinds of bounded noise models are classified. For the noise energy bounded case, the observability ellipsoid and the image observability ellipsoid are defined by the state observed error and then a novel concept on the LDT systems, called as the observe ability, is proposed. Based on that, some theorems and properties about the observe ability and the signal detecting performances are given and proven, and then the reason that to maximize the observe ability is to optimize the signal detecting performances is established. Secondly, a dual relation between the observability ellipsoid and the controllability ellipsoid, which volumes and radii are respectively with some inverse relations, is stated and proven. Accordingly, the analytical computing equations for the volume of the two observability ellipsoids are got and some analytical shape factors of these ellipsoids are deconstructed. Based on these effective compting for the volumes, radii, and shape factors, analyzing and optimizing for the observe ability can be carried out. Thirdly, to compare rationally the state observe ability between the different systems or different system parameters, the normalization of the output variables, the state variables, and the system models are discussed. Finally, some numerical experiments and their results show the effectiveness of the computing and comparing methods for the observe ability.

false

true

false

227,063

2207.14373

Eye Gaze Estimation Model Analysis

We explore techniques for eye gaze estimation using machine learning. Eye gaze estimation is a common problem for various behavior analysis and human-computer interfaces. The purpose of this work is to discuss various model types for eye gaze estimation and present the results from predicting gaze direction using eye landmarks in unconstrained settings. In unconstrained real-world settings, feature-based and model-based methods are outperformed by recent appearance-based methods due to factors like illumination changes and other visual artifacts. We discuss a learning-based method for eye region landmark localization trained exclusively on synthetic data. We discuss how to use detected landmarks as input to iterative model-fitting and lightweight learning-based gaze estimation methods and how to use the model for person-independent and personalized gaze estimations.

false

true

false

true

false

310,558

2305.16360

Modeling Task Relationships in Multi-variate Soft Sensor with Balanced Mixture-of-Experts

Accurate estimation of multiple quality variables is critical for building industrial soft sensor models, which have long been confronted with data efficiency and negative transfer issues. Methods sharing backbone parameters among tasks address the data efficiency issue; however, they still fail to mitigate the negative transfer problem. To address this issue, a balanced Mixture-of-Experts (BMoE) is proposed in this work, which consists of a multi-gate mixture of experts (MMoE) module and a task gradient balancing (TGB) module. The MoE module aims to portray task relationships, while the TGB module balances the gradients among tasks dynamically. Both of them cooperate to mitigate the negative transfer problem. Experiments on the typical sulfur recovery unit demonstrate that BMoE models task relationship and balances the training process effectively, and achieves better performance than baseline models significantly.

false

true

false

true

false

368,051

2209.09702

LEMURS: Learning Distributed Multi-Robot Interactions

This paper presents LEMURS, an algorithm for learning scalable multi-robot control policies from cooperative task demonstrations. We propose a port-Hamiltonian description of the multi-robot system to exploit universal physical constraints in interconnected systems and achieve closed-loop stability. We represent a multi-robot control policy using an architecture that combines self-attention mechanisms and neural ordinary differential equations. The former handles time-varying communication in the robot team, while the latter respects the continuous-time robot dynamics. Our representation is distributed by construction, enabling the learned control policies to be deployed in robot teams of different sizes. We demonstrate that LEMURS can learn interactions and cooperative behaviors from demonstrations of multi-agent navigation and flocking tasks.

false

true

false

true

false

318,616

0811.0146

Effect of Tuned Parameters on a LSA MCQ Answering Model

This paper presents the current state of a work in progress, whose objective is to better understand the effects of factors that significantly influence the performance of Latent Semantic Analysis (LSA). A difficult task, which consists in answering (French) biology Multiple Choice Questions, is used to test the semantic properties of the truncated singular space and to study the relative influence of main parameters. A dedicated software has been designed to fine tune the LSA semantic space for the Multiple Choice Questions task. With optimal parameters, the performances of our simple model are quite surprisingly equal or superior to those of 7th and 8th grades students. This indicates that semantic spaces were quite good despite their low dimensions and the small sizes of training data sets. Besides, we present an original entropy global weighting of answers' terms of each question of the Multiple Choice Questions which was necessary to achieve the model's success.

false

true

false

true

false

2,601

1410.7026

Optimal topology of multi-agent systems with two leaders: a zero-sum game perspective

It is typical to assume that there is no conflict of interest among leaders. Under such assumption, it is known that, for a multi-agent system with two leaders, if the followers' interaction subgraph is undirected and connected, then followers will converge to a convex combination of two leaders' states with linear consensus protocol. In this paper, we introduce the conflict between leaders: by choosing k followers to connect with, every leader attempts all followers converge to himself closer than that of the other. By using graph theory and matrix theory, we formulate this conflict as a standard two-player zero-sum game and give some properties about it. It is noteworthy that the interaction graph here is generated from the conflict between leaders. Interestingly, we find that to find the optimal topology of the system is equivalent to solve a Nash equilibrium. Especially for the case of choosing one connected follower, the necessary and sufficient condition for an interaction graph to be the optimal one is given. Moreover, if followers' interaction graph is a circulant graph or a graph with a center node, then the system's optimal topology is obtained. Simulation examples are provided to validate the effectiveness of the theoretical results.

false

true

false

true

false

true

37,032

1209.2541

Absence of epidemic thresholds in a growing adaptive network

The structure of social contact networks strongly influences the dynamics of epidemic diseases. In particular the scale-free structure of real-world social networks allows unlikely diseases with low infection rates to spread and become endemic. However, in particular for potentially fatal diseases, also the impact of the disease on the social structure cannot be neglected, leading to a complex interplay. Here, we consider the growth of a network by preferential attachment from which nodes are simultaneously removed due to an SIR epidemic. We show that increased infectiousness increases the prevalence of the disease and simultaneously causes a transition from scale-free to exponential topology. Although a transition to a degree distribution with finite variance takes place, the network still exhibits no epidemic threshold in the thermodynamic limit. We illustrate these results using agent-based simulations and analytically tractable approximation schemes.

false

true

false

18,516

2011.07577

Placement in Integrated Circuits using Cyclic Reinforcement Learning and Simulated Annealing

Physical design and production of Integrated Circuits (IC) is becoming increasingly more challenging as the sophistication in IC technology is steadily increasing. Placement has been one of the most critical steps in IC physical design. Through decades of research, partition-based, analytical-based and annealing-based placers have been enriching the placement solution toolbox. However, open challenges including long run time and lack of ability to generalize continue to restrict wider applications of existing placement tools. We devise a learning-based placement tool based on cyclic application of Reinforcement Learning (RL) and Simulated Annealing (SA) by leveraging the advancement of RL. Results show that the RL module is able to provide a better initialization for SA and thus leads to a better final placement design. Compared to other recent learning-based placers, our method is majorly different with its combination of RL and SA. It leverages the RL model's ability to quickly get a good rough solution after training and the heuristic's ability to realize greedy improvements in the solution.

false

true

false

true

false

206,602

2406.06014

Network two-sample test for block models

We consider the two-sample testing problem for networks, where the goal is to determine whether two sets of networks originated from the same stochastic model. Assuming no vertex correspondence and allowing for different numbers of nodes, we address a fundamental network testing problem that goes beyond simple adjacency matrix comparisons. We adopt the stochastic block model (SBM) for network distributions, due to their interpretability and the potential to approximate more general models. The lack of meaningful node labels and vertex correspondence translate to a graph matching challenge when developing a test for SBMs. We introduce an efficient algorithm to match estimated network parameters, allowing us to properly combine and contrast information within and across samples, leading to a powerful test. We show that the matching algorithm, and the overall test are consistent, under mild conditions on the sparsity of the networks and the sample sizes, and derive a chi-squared asymptotic null distribution for the test. Through a mixture of theoretical insights and empirical validations, including experiments with both synthetic and real-world data, this study advances robust statistical inference for complex network data.

false

true

false

462,396

2208.10739

Quality-Constant Per-Shot Encoding by Two-Pass Learning-based Rate Factor Prediction

Providing quality-constant streams can simultaneously guarantee user experience and prevent wasting bit-rate. In this paper, we propose a novel deep learning based two-pass encoder parameter prediction framework to decide rate factor (RF), with which encoder can output streams with constant quality. For each one-shot segment in a video, the proposed method firstly extracts spatial, temporal and pre-coding features by an ultra fast pre-process. Based on these features, a RF parameter is predicted by a deep neural network. Video encoder uses the RF to compress segment as the first encoding pass. Then VMAF quality of the first pass encoding is measured. If the quality doesn't meet target, a second pass RF prediction and encoding will be performed. With the help of first pass predicted RF and corresponding actual quality as feedback, the second pass prediction will be highly accurate. Experiments show the proposed method requires only 1.55 times encoding complexity on average, meanwhile the accuracy, that the compressed video's actual VMAF is within $\pm1$ around the target VMAF, reaches 98.88%.

false

true

false

314,181

2404.09453

Towards Greener Nights: Exploring AI-Driven Solutions for Light Pollution Management

This research endeavors to address the pervasive issue of light pollution through an interdisciplinary approach, leveraging data science and machine learning techniques. By analyzing extensive datasets and research findings, we aim to develop predictive models capable of estimating the degree of sky glow observed in various locations and times. Our research seeks to inform evidence-based interventions and promote responsible outdoor lighting practices to mitigate the adverse impacts of light pollution on ecosystems, energy consumption, and human well-being.

false

true

false

446,676

1906.05838

Goal-conditioned Imitation Learning

Designing rewards for Reinforcement Learning (RL) is challenging because it needs to convey the desired task, be efficient to optimize, and be easy to compute. The latter is particularly problematic when applying RL to robotics, where detecting whether the desired configuration is reached might require considerable supervision and instrumentation. Furthermore, we are often interested in being able to reach a wide range of configurations, hence setting up a different reward every time might be unpractical. Methods like Hindsight Experience Replay (HER) have recently shown promise to learn policies able to reach many goals, without the need of a reward. Unfortunately, without tricks like resetting to points along the trajectory, HER might require many samples to discover how to reach certain areas of the state-space. In this work we investigate different approaches to incorporate demonstrations to drastically speed up the convergence to a policy able to reach any goal, also surpassing the performance of an agent trained with other Imitation Learning algorithms. Furthermore, we show our method can also be used when the available expert trajectories do not contain the actions, which can leverage kinesthetic or third person demonstration. The code is available at https://sites.google.com/view/goalconditioned-il/.

false

true

false

true

false

true

false

135,136

2403.11057

Large Language Models Powered Context-aware Motion Prediction in Autonomous Driving

Motion prediction is among the most fundamental tasks in autonomous driving. Traditional methods of motion forecasting primarily encode vector information of maps and historical trajectory data of traffic participants, lacking a comprehensive understanding of overall traffic semantics, which in turn affects the performance of prediction tasks. In this paper, we utilized Large Language Models (LLMs) to enhance the global traffic context understanding for motion prediction tasks. We first conducted systematic prompt engineering, visualizing complex traffic environments and historical trajectory information of traffic participants into image prompts -- Transportation Context Map (TC-Map), accompanied by corresponding text prompts. Through this approach, we obtained rich traffic context information from the LLM. By integrating this information into the motion prediction model, we demonstrate that such context can enhance the accuracy of motion predictions. Furthermore, considering the cost associated with LLMs, we propose a cost-effective deployment strategy: enhancing the accuracy of motion prediction tasks at scale with 0.7\% LLM-augmented datasets. Our research offers valuable insights into enhancing the understanding of traffic scenes of LLMs and the motion prediction performance of autonomous driving. The source code is available at \url{https://github.com/AIR-DISCOVER/LLM-Augmented-MTR} and \url{https://aistudio.baidu.com/projectdetail/7809548}.

false

true

false

true

false

438,503

2403.02454

The Ink Splotch Effect: A Case Study on ChatGPT as a Co-Creative Game Designer

This paper studies how large language models (LLMs) can act as effective, high-level creative collaborators and ``muses'' for game design. We model the design of this study after the exercises artists use by looking at amorphous ink splotches for creative inspiration. Our goal is to determine whether AI-assistance can improve, hinder, or provide an alternative quality to games when compared to the creative intents implemented by human designers. The capabilities of LLMs as game designers are stress tested by placing it at the forefront of the decision making process. Three prototype games are designed across 3 different genres: (1) a minimalist base game, (2) a game with features and game feel elements added by a human game designer, and (3) a game with features and feel elements directly implemented from prompted outputs of the LLM, ChatGPT. A user study was conducted and participants were asked to blindly evaluate the quality and their preference of these games. We discuss both the development process of communicating creative intent to an AI chatbot and the synthesized open feedback of the participants. We use this data to determine both the benefits and shortcomings of AI in a more design-centric role.

false

true

false

434,808

2501.13134

UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior

Image restoration aims to recover content from inputs degraded by various factors, such as adverse weather, blur, and noise. Perceptual Image Restoration (PIR) methods improve visual quality but often do not support downstream tasks effectively. On the other hand, Task-oriented Image Restoration (TIR) methods focus on enhancing image utility for high-level vision tasks, sometimes compromising visual quality. This paper introduces UniRestore, a unified image restoration model that bridges the gap between PIR and TIR by using a diffusion prior. The diffusion prior is designed to generate images that align with human visual quality preferences, but these images are often unsuitable for TIR scenarios. To solve this limitation, UniRestore utilizes encoder features from an autoencoder to adapt the diffusion prior to specific tasks. We propose a Complementary Feature Restoration Module (CFRM) to reconstruct degraded encoder features and a Task Feature Adapter (TFA) module to facilitate adaptive feature fusion in the decoder. This design allows UniRestore to optimize images for both human perception and downstream task requirements, addressing discrepancies between visual quality and functional needs. Integrating these modules also enhances UniRestore's adapability and efficiency across diverse tasks. Extensive expertments demonstrate the superior performance of UniRestore in both PIR and TIR scenarios.

false

true

false

526,575

2407.02953

Affine Frequency Division Multiplexing for Compressed Sensing of Time-Varying Channels

This paper addresses compressed sensing of linear time-varying (LTV) wireless propagation links under the assumption of double sparsity i.e., sparsity in both the delay and Doppler domains, using Affine Frequency Division Multiplexing (AFDM) measurements. By rigorously linking the double sparsity model to the hierarchical sparsity paradigm, a compressed sensing algorithm with recovery guarantees is proposed for extracting delay-Doppler profiles of LTV channels using AFDM. Through mathematical analysis and numerical results, the superiority of AFDM over other waveforms in terms of channel estimation overhead and minimal sampling rate requirements in sub-Nyquist radar applications is demonstrated.

false

true

false

469,960

2003.10959

Learning to Exploit Multiple Vision Modalities by Using Grafted Networks

Novel vision sensors such as thermal, hyperspectral, polarization, and event cameras provide information that is not available from conventional intensity cameras. An obstacle to using these sensors with current powerful deep neural networks is the lack of large labeled training datasets. This paper proposes a Network Grafting Algorithm (NGA), where a new front end network driven by unconventional visual inputs replaces the front end network of a pretrained deep network that processes intensity frames. The self-supervised training uses only synchronously-recorded intensity frames and novel sensor data to maximize feature similarity between the pretrained network and the grafted network. We show that the enhanced grafted network reaches competitive average precision (AP50) scores to the pretrained network on an object detection task using thermal and event camera datasets, with no increase in inference costs. Particularly, the grafted network driven by thermal frames showed a relative improvement of 49.11% over the use of intensity frames. The grafted front end has only 5--8% of the total parameters and can be trained in a few hours on a single GPU equivalent to 5% of the time that would be needed to train the entire object detector from labeled data. NGA allows new vision sensors to capitalize on previously pretrained powerful deep models, saving on training cost and widening a range of applications for novel sensors.

false

true

false

true

false

169,478

2104.06689

Learning Normal Dynamics in Videos with Meta Prototype Network

Frame reconstruction (current or future frame) based on Auto-Encoder (AE) is a popular method for video anomaly detection. With models trained on the normal data, the reconstruction errors of anomalous scenes are usually much larger than those of normal ones. Previous methods introduced the memory bank into AE, for encoding diverse normal patterns across the training videos. However, they are memory-consuming and cannot cope with unseen new scenarios in the testing data. In this work, we propose a dynamic prototype unit (DPU) to encode the normal dynamics as prototypes in real time, free from extra memory cost. In addition, we introduce meta-learning to our DPU to form a novel few-shot normalcy learner, namely Meta-Prototype Unit (MPU). It enables the fast adaption capability on new scenes by only consuming a few iterations of update. Extensive experiments are conducted on various benchmarks. The superior performance over the state-of-the-art demonstrates the effectiveness of our method.

false

true

false

230,161

2311.05840

Predictive AI for SME and Large Enterprise Financial Performance Management

Financial performance management is at the core of business management and has historically relied on financial ratio analysis using Balance Sheet and Income Statement data to assess company performance as compared with competitors. Little progress has been made in predicting how a company will perform or in assessing the risks (probabilities) of financial underperformance. In this study I introduce a new set of financial and macroeconomic ratios that supplement standard ratios of Balance Sheet and Income Statement. I also provide a set of supervised learning models (ML Regressors and Neural Networks) and Bayesian models to predict company performance. I conclude that the new proposed variables improve model accuracy when used in tandem with standard industry ratios. I also conclude that Feedforward Neural Networks (FNN) are simpler to implement and perform best across 6 predictive tasks (ROA, ROE, Net Margin, Op Margin, Cash Ratio and Op Cash Generation); although Bayesian Networks (BN) can outperform FNN under very specific conditions. BNs have the additional benefit of providing a probability density function in addition to the predicted (expected) value. The study findings have significant potential helping CFOs and CEOs assess risks of financial underperformance to steer companies in more profitable directions; supporting lenders in better assessing the condition of a company and providing investors with tools to dissect financial statements of public companies more accurately.

false

true

false

406,724

2011.05533

Spoken Language Interaction with Robots: Research Issues and Recommendations, Report from the NSF Future Directions Workshop

With robotics rapidly advancing, more effective human-robot interaction is increasingly needed to realize the full potential of robots for society. While spoken language must be part of the solution, our ability to provide spoken language interaction capabilities is still very limited. The National Science Foundation accordingly convened a workshop, bringing together speech, language, and robotics researchers to discuss what needs to be done. The result is this report, in which we identify key scientific and engineering advances needed. Our recommendations broadly relate to eight general themes. First, meeting human needs requires addressing new challenges in speech technology and user experience design. Second, this requires better models of the social and interactive aspects of language use. Third, for robustness, robots need higher-bandwidth communication with users and better handling of uncertainty, including simultaneous consideration of multiple hypotheses and goals. Fourth, more powerful adaptation methods are needed, to enable robots to communicate in new environments, for new tasks, and with diverse user populations, without extensive re-engineering or the collection of massive training data. Fifth, since robots are embodied, speech should function together with other communication modalities, such as gaze, gesture, posture, and motion. Sixth, since robots operate in complex environments, speech components need access to rich yet efficient representations of what the robot knows about objects, locations, noise sources, the user, and other humans. Seventh, since robots operate in real time, their speech and language processing components must also. Eighth, in addition to more research, we need more work on infrastructure and resources, including shareable software modules and internal interfaces, inexpensive hardware, baseline systems, and diverse corpora.

true

false

true

false

205,951

2406.10719

Trading Devil: Robust backdoor attack via Stochastic investment models and Bayesian approach

With the growing use of voice-activated systems and speech recognition technologies, the danger of backdoor attacks on audio data has grown significantly. This research looks at a specific type of attack, known as a Stochastic investment-based backdoor attack (MarketBack), in which adversaries strategically manipulate the stylistic properties of audio to fool speech recognition systems. The security and integrity of machine learning models are seriously threatened by backdoor attacks, in order to maintain the reliability of audio applications and systems, the identification of such attacks becomes crucial in the context of audio data. Experimental results demonstrated that MarketBack is feasible to achieve an average attack success rate close to 100% in seven victim models when poisoning less than 1% of the training data.

false

true

false

true

false

464,522

1909.01136

Pre-training A Neural Language Model Improves The Sample Efficiency of an Emergency Room Classification Model

To build a French national electronic injury surveillance system based on emergency room visits, we aim to develop a coding system to classify their causes from clinical notes in free-text. Supervised learning techniques have shown good results in this area but require a large amount of expert annotated dataset which is time consuming and costly to obtain. We hypothesize that the Natural Language Processing Transformer model incorporating a generative self-supervised pre-training step can significantly reduce the required number of annotated samples for supervised fine-tuning. In this preliminary study, we test our hypothesis in the simplified problem of predicting whether a visit is the consequence of a traumatic event or not from free-text clinical notes. Using fully re-trained GPT-2 models (without OpenAI pre-trained weights), we assess the gain of applying a self-supervised pre-training phase with unlabeled notes prior to the supervised learning task. Results show that the number of data required to achieve a ginve level of performance (AUC>0.95) was reduced by a factor of 10 when applying pre-training. Namely, for 16 times more data, the fully-supervised model achieved an improvement <1% in AUC. To conclude, it is possible to adapt a multi-purpose neural language model such as the GPT-2 to create a powerful tool for classification of free-text notes with only a small number of labeled samples.

false

true

false

true

false

true

false

143,822

2410.19110

Bio2Token: All-atom tokenization of any biomolecular structure with Mamba

Efficient encoding and representation of large 3D molecular structures with high fidelity is critical for biomolecular design applications. Despite this, many representation learning approaches restrict themselves to modeling smaller systems or use coarse-grained approximations of the systems, for example modeling proteins at the resolution of amino acid residues rather than at the level of individual atoms. To address this, we develop quantized auto-encoders that learn atom-level tokenizations of complete proteins, RNA and small molecule structures with reconstruction accuracies well below 1 Angstrom. We demonstrate that a simple Mamba state space model architecture is efficient compared to an SE(3)-invariant IPA architecture, reaches competitive accuracies and can scale to systems with almost 100,000 atoms. The learned structure tokens of bio2token may serve as the input for all-atom generative models in the future.

false

true

false

true

false

502,160

2003.03229

Non-linear Neurons with Human-like Apical Dendrite Activations

In order to classify linearly non-separable data, neurons are typically organized into multi-layer neural networks that are equipped with at least one hidden layer. Inspired by some recent discoveries in neuroscience, we propose a new model of artificial neuron along with a novel activation function enabling the learning of nonlinear decision boundaries using a single neuron. We show that a standard neuron followed by our novel apical dendrite activation (ADA) can learn the XOR logical function with 100% accuracy. Furthermore, we conduct experiments on six benchmark data sets from computer vision, signal processing and natural language processing, i.e. MOROCO, UTKFace, CREMA-D, Fashion-MNIST, Tiny ImageNet and ImageNet, showing that the ADA and the leaky ADA functions provide superior results to Rectified Linear Units (ReLU), leaky ReLU, RBF and Swish, for various neural network architectures, e.g. one-hidden-layer or two-hidden-layer multi-layer perceptrons (MLPs) and convolutional neural networks (CNNs) such as LeNet, VGG, ResNet and Character-level CNN. We obtain further performance improvements when we change the standard model of the neuron with our pyramidal neuron with apical dendrite activations (PyNADA). Our code is available at: https://github.com/raduionescu/pynada.

false

true

false

true

false

true

false

167,164

2207.10541

Unveiling the Latent Space Geometry of Push-Forward Generative Models

Many deep generative models are defined as a push-forward of a Gaussian measure by a continuous generator, such as Generative Adversarial Networks (GANs) or Variational Auto-Encoders (VAEs). This work explores the latent space of such deep generative models. A key issue with these models is their tendency to output samples outside of the support of the target distribution when learning disconnected distributions. We investigate the relationship between the performance of these models and the geometry of their latent space. Building on recent developments in geometric measure theory, we prove a sufficient condition for optimality in the case where the dimension of the latent space is larger than the number of modes. Through experiments on GANs, we demonstrate the validity of our theoretical results and gain new insights into the latent space geometry of these models. Additionally, we propose a truncation method that enforces a simplicial cluster structure in the latent space and improves the performance of GANs.

false

true

false

true

false

309,297

2107.08964

Transductive image segmentation: Self-training and effect of uncertainty estimation

Semi-supervised learning (SSL) uses unlabeled data during training to learn better models. Previous studies on SSL for medical image segmentation focused mostly on improving model generalization to unseen data. In some applications, however, our primary interest is not generalization but to obtain optimal predictions on a specific unlabeled database that is fully available during model development. Examples include population studies for extracting imaging phenotypes. This work investigates an often overlooked aspect of SSL, transduction. It focuses on the quality of predictions made on the unlabeled data of interest when they are included for optimization during training, rather than improving generalization. We focus on the self-training framework and explore its potential for transduction. We analyze it through the lens of Information Gain and reveal that learning benefits from the use of calibrated or under-confident models. Our extensive experiments on a large MRI database for multi-class segmentation of traumatic brain lesions shows promising results when comparing transductive with inductive predictions. We believe this study will inspire further research on transductive learning, a well-suited paradigm for medical image analysis.

false

true

false

246,892

2402.09050

End-to-End Training Induces Information Bottleneck through Layer-Role Differentiation: A Comparative Analysis with Layer-wise Training

End-to-end (E2E) training, optimizing the entire model through error backpropagation, fundamentally supports the advancements of deep learning. Despite its high performance, E2E training faces the problems of memory consumption, parallel computing, and discrepancy with the functionalities of the actual brain. Various alternative methods have been proposed to overcome these difficulties; however, no one can yet match the performance of E2E training, thereby falling short in practicality. Furthermore, there is no deep understanding regarding differences in the trained model properties beyond the performance gap. In this paper, we reconsider why E2E training demonstrates a superior performance through a comparison with layer-wise training, a non-E2E method that locally sets errors. On the basis of the observation that E2E training has an advantage in propagating input information, we analyze the information plane dynamics of intermediate representations based on the Hilbert-Schmidt independence criterion (HSIC). The results of our normalized HSIC value analysis reveal the E2E training ability to exhibit different information dynamics across layers, in addition to efficient information propagation. Furthermore, we show that this layer-role differentiation leads to the final representation following the information bottleneck principle. It suggests the need to consider the cooperative interactions between layers, not just the final layer when analyzing the information bottleneck of deep learning.

false

true

false

429,344

2004.11587

Concept Drift Detection via Equal Intensity k-means Space Partitioning

Data stream poses additional challenges to statistical classification tasks because distributions of the training and target samples may differ as time passes. Such distribution change in streaming data is called concept drift. Numerous histogram-based distribution change detection methods have been proposed to detect drift. Most histograms are developed on grid-based or tree-based space partitioning algorithms which makes the space partitions arbitrary, unexplainable, and may cause drift blind-spots. There is a need to improve the drift detection accuracy for histogram-based methods with the unsupervised setting. To address this problem, we propose a cluster-based histogram, called equal intensity k-means space partitioning (EI-kMeans). In addition, a heuristic method to improve the sensitivity of drift detection is introduced. The fundamental idea of improving the sensitivity is to minimize the risk of creating partitions in distribution offset regions. Pearson's chi-square test is used as the statistical hypothesis test so that the test statistics remain independent of the sample distribution. The number of bins and their shapes, which strongly influence the ability to detect drift, are determined dynamically from the sample based on an asymptotic constraint in the chi-square test. Accordingly, three algorithms are developed to implement concept drift detection, including a greedy centroids initialization algorithm, a cluster amplify-shrink algorithm, and a drift detection algorithm. For drift adaptation, we recommend retraining the learner if a drift is detected. The results of experiments on synthetic and real-world datasets demonstrate the advantages of EI-kMeans and show its efficacy in detecting concept drift.

false

true

false

173,953

2206.01815

Option Discovery for Autonomous Generation of Symbolic Knowledge

In this work we present an empirical study where we demonstrate the possibility of developing an artificial agent that is capable to autonomously explore an experimental scenario. During the exploration, the agent is able to discover and learn interesting options allowing to interact with the environment without any pre-assigned goal, then abstract and re-use the acquired knowledge to solve possible tasks assigned ex-post. We test the system in the so-called Treasure Game domain described in the recent literature and we empirically demonstrate that the discovered options can be abstracted in an probabilistic symbolic planning model (using the PPDDL language), which allowed the agent to generate symbolic plans to achieve extrinsic goals.

false

true

false

300,613

2206.13028

Multi-Scale Spatial Temporal Graph Convolutional Network for Skeleton-Based Action Recognition

Graph convolutional networks have been widely used for skeleton-based action recognition due to their excellent modeling ability of non-Euclidean data. As the graph convolution is a local operation, it can only utilize the short-range joint dependencies and short-term trajectory but fails to directly model the distant joints relations and long-range temporal information that are vital to distinguishing various actions. To solve this problem, we present a multi-scale spatial graph convolution (MS-GC) module and a multi-scale temporal graph convolution (MT-GC) module to enrich the receptive field of the model in spatial and temporal dimensions. Concretely, the MS-GC and MT-GC modules decompose the corresponding local graph convolution into a set of sub-graph convolution, forming a hierarchical residual architecture. Without introducing additional parameters, the features will be processed with a series of sub-graph convolutions, and each node could complete multiple spatial and temporal aggregations with its neighborhoods. The final equivalent receptive field is accordingly enlarged, which is capable of capturing both short- and long-range dependencies in spatial and temporal domains. By coupling these two modules as a basic block, we further propose a multi-scale spatial temporal graph convolutional network (MST-GCN), which stacks multiple blocks to learn effective motion representations for action recognition. The proposed MST-GCN achieves remarkable performance on three challenging benchmark datasets, NTU RGB+D, NTU-120 RGB+D and Kinetics-Skeleton, for skeleton-based action recognition.

false

true

false

304,811

1406.0234

Joint PIC and relay selection based on greedy techniques for cooperative DS-CDMA systems

In this work, we propose a cross-layer design strategy based on the parallel interference cancellation (PIC) detection technique and a multi-relay selection algorithm for the uplink of cooperative direct-sequence code-division multiple access (DS-CDMA) systems. We devise a low-cost greedy list-based PIC (GL-PIC) strategy with RAKE receivers as the front-end that can approach the maximum likelihood detector performance. We also present a low-complexity multi-relay selection algorithm based on greedy techniques that can approach the performance of an exhaustive search. Simulations show an excellent bit error rate performance of the proposed detection and relay selection algorithms as compared to existing techniques.

false

true

false

33,542

1905.08846

Discovering Hidden Structure in High Dimensional Human Behavioral Data via Tensor Factorization

In recent years, the rapid growth in technology has increased the opportunity for longitudinal human behavioral studies. Rich multimodal data, from wearables like Fitbit, online social networks, mobile phones etc. can be collected in natural environments. Uncovering the underlying low-dimensional structure of noisy multi-way data in an unsupervised setting is a challenging problem. Tensor factorization has been successful in extracting the interconnected low-dimensional descriptions of multi-way data. In this paper, we apply non-negative tensor factorization on a real-word wearable sensor data, StudentLife, to find latent temporal factors and group of similar individuals. Meta data is available for the semester schedule, as well as the individuals' performance and personality. We demonstrate that non-negative tensor factorization can successfully discover clusters of individuals who exhibit higher academic performance, as well as those who frequently engage in leisure activities. The recovered latent temporal patterns associated with these groups are validated against ground truth data to demonstrate the accuracy of our framework.

false

true

false

131,581

2310.06956

Adversarial optimization leads to over-optimistic security-constrained dispatch, but sampling can help

To ensure safe, reliable operation of the electrical grid, we must be able to predict and mitigate likely failures. This need motivates the classic security-constrained AC optimal power flow (SCOPF) problem. SCOPF is commonly solved using adversarial optimization, where the dispatcher and an adversary take turns optimizing a robust dispatch and adversarial attack, respectively. We show that adversarial optimization is liable to severely overestimate the robustness of the optimized dispatch (when the adversary encounters a local minimum), leading the operator to falsely believe that their dispatch is secure. To prevent this overconfidence, we develop a novel adversarial sampling approach that prioritizes diversity in the predicted attacks. We find that our method not only substantially improves the robustness of the optimized dispatch but also avoids overconfidence, accurately characterizing the likelihood of voltage collapse under a given threat model. We demonstrate a proof-of-concept on small-scale transmission systems with 14 and 57 nodes.

false

true

false

398,774

1907.02865

Cardiac MRI Segmentation with Strong Anatomical Guarantees

Recent publications have shown that the segmentation accuracy of modern-day convolutional neural networks (CNN) applied on cardiac MRI can reach the inter-expert variability, a great achievement in this area of research. However, despite these successes, CNNs still produce anatomically inaccurate segmentations as they provide no guarantee on the anatomical plausibility of their outcome, even when using a shape prior. In this paper, we propose a cardiac MRI segmentation method which always produces anatomically plausible results. At the core of the method is an adversarial variational autoencoder (aVAE) whose latent space encodes a smooth manifold on which lies a large spectrum of valid cardiac shapes. This aVAE is used to automatically warp anatomically inaccurate cardiac shapes towards a close but correct shape. Our method can accommodate any cardiac segmentation method and convert its anatomically implausible results to plausible ones without affecting its overall geometric and clinical metrics. With our method, CNNs can now produce results that are both within the inter-expert variability and always anatomically plausible.

false

true

false

137,705

2407.20174

Advancing Multimodal Large Language Models in Chart Question Answering with Visualization-Referenced Instruction Tuning

Emerging multimodal large language models (MLLMs) exhibit great potential for chart question answering (CQA). Recent efforts primarily focus on scaling up training datasets (i.e., charts, data tables, and question-answer (QA) pairs) through data collection and synthesis. However, our empirical study on existing MLLMs and CQA datasets reveals notable gaps. First, current data collection and synthesis focus on data volume and lack consideration of fine-grained visual encodings and QA tasks, resulting in unbalanced data distribution divergent from practical CQA scenarios. Second, existing work follows the training recipe of the base MLLMs initially designed for natural images, under-exploring the adaptation to unique chart characteristics, such as rich text elements. To fill the gap, we propose a visualization-referenced instruction tuning approach to guide the training dataset enhancement and model development. Specifically, we propose a novel data engine to effectively filter diverse and high-quality data from existing datasets and subsequently refine and augment the data using LLM-based generation techniques to better align with practical QA tasks and visual encodings. Then, to facilitate the adaptation to chart characteristics, we utilize the enriched data to train an MLLM by unfreezing the vision encoder and incorporating a mixture-of-resolution adaptation strategy for enhanced fine-grained recognition. Experimental results validate the effectiveness of our approach. Even with fewer training examples, our model consistently outperforms state-of-the-art CQA models on established benchmarks. We also contribute a dataset split as a benchmark for future research. Source codes and datasets of this paper are available at https://github.com/zengxingchen/ChartQA-MLLM.

false

true

false

true

false

477,070