paper_id stringlengths 3 14 | source stringclasses 3
values | title stringlengths 0 287 ⌀ | abstract stringlengths 84 3.8k | num_figures int32 0 2 | paper_json stringlengths 1.13k 162k | method_figure imagewidth (px) 246 1.69k ⌀ | result_figure imagewidth (px) 248 1.7k ⌀ |
|---|---|---|---|---|---|---|---|
749 | iclr_2017 | DEEP CONVOLUTIONAL NEURAL NETWORK DESIGN PATTERNS | Recent research in the deep learning field has produced a plethora of new architectures. At the same time, a growing number of groups are applying deep learning to new applications. Some of these groups are likely to be composed of inexperienced deep learning practitioners who are baffled by the dizzying array of archi... | 0 | {"name": "749.pdf", "metadata": {"source": "CRF", "title": "DEEP CONVOLUTIONAL NEURAL NETWORK DESIGN PATTERNS", "authors": ["Leslie N. Smith", "Nicholay Topin"], "emails": ["leslie.smith@nrl.navy.mil", "ntopin1@umbc.edu"], "sections": [{"heading": "1 INTRODUCTION", "text": "Many recent articles discuss new architecture... | Not supported with pagination yet | Not supported with pagination yet |
750 | iclr_2017 | null | Low-shot visual learning, the ability to recognize novel object categories from very few, or even one example, is a hallmark of human visual intelligence. Though successful on many tasks, deep learning approaches tends to be notoriously datahungry. Recently, feature penalty regularization has been proved effective on c... | 1 | {"name": "750.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["LOW-SHOT LEARNING", "Zhuoyuan Chen", "Xiao Liu", "Wei Xu"], "emails": ["chenzhuoyuan@baidu.com", "liuxiao@baidu.com", "wei.xu@baidu.com", "han.zhao@cs.cmu.edu"], "sections": [{"heading": "1 INTRODUCTION", "text": "The current success of deep ... | Not supported with pagination yet | |
751 | iclr_2017 | COMBATING DEEP REINFORCEMENT LEARNING’S SISYPHEAN CURSE WITH INTRINSIC FEAR | To use deep reinforcement learning in the wild, we might hope for an agent that can avoid catastrophic mistakes. Unfortunately, even in simple environments, the popular deep Q-network (DQN) algorithm is doomed by a Sisyphean curse. Owing to the use of function approximation, these agents eventually forget experiences a... | 1 | {"name": "751.pdf", "metadata": {"source": "CRF", "title": "COMBATING DEEP REINFORCEMENT LEARNING’S SISYPHEAN CURSE WITH INTRINSIC FEAR", "authors": ["Zachary C. Lipton", "Jianfeng Gao", "Lihong Li", "Jianshu Chen", "Li Deng"], "emails": ["zlipton@cs.ucsd.edu", "deng}@microsoft.com"], "sections": [{"heading": "1 INTROD... | Not supported with pagination yet | |
754 | iclr_2017 | LEARNING PYTHON CODE SUGGESTION WITH A SPARSE POINTER NETWORK | To enhance developer productivity, all modern integrated development environments (IDEs) include code suggestion functionality that proposes likely next tokens at the cursor. While current IDEs work well for statically-typed languages, their reliance on type annotations means that they do not provide the same level of ... | 2 | {"name": "754.pdf", "metadata": {"source": "CRF", "title": "LEARNING PYTHON CODE SUGGESTION WITH A SPARSE POINTER NETWORK", "authors": ["Avishkar Bhoopchand", "Tim Rocktäschel", "Earl Barr", "Sebastian Riedel"], "emails": ["avishkar.bhoopchand.15@ucl.ac.uk,", "t.rocktaschel@cs.ucl.ac.uk", "e.barr@cs.ucl.ac.uk", "s.ried... | ||
755 | iclr_2017 | null | We provide a theoretical explanation for the great performance of ResNet via the study of deep linear networks and some nonlinear variants. We show that with or without nonlinearities, by adding shortcuts that have depth two, the condition number of the Hessian of the loss function at the zero initial point is depthinv... | 1 | {"name": "755.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["DEMYSTIFYING RESNET", "Sihan Li", "Jiantao Jiao", "Yanjun Han", "Tsachy Weissman"], "emails": ["lisihan13@mails.tsinghua.edu.cn", "jiantao@stanford.edu", "yjhan@stanford.edu", "tsachy@stanford.edu"], "sections": [{"heading": "1 INTRODUCTION",... | Not supported with pagination yet | |
756 | iclr_2017 | TIONS WITH NEURAL SIMILARITY AND CONTEXT EN- CODERS | We introduce similarity encoders (SimEc), which learn similarity preserving representations by using a feed-forward neural network to map data into an embedding space where the original similarities can be approximated linearly. The model can easily compute representations for novel (out-of-sample) data points, even if... | 0 | {"name": "756.pdf", "metadata": {"source": "CRF", "title": "TIONS WITH NEURAL SIMILARITY AND CONTEXT EN- CODERS", "authors": ["Franziska Horn"], "emails": ["franziska.horn@campus.tu-berlin.de", "klaus-robert.mueller@tu-berlin.de"], "sections": [{"heading": "1 INTRODUCTION", "text": "Many dimensionality reduction or man... | Not supported with pagination yet | Not supported with pagination yet |
760 | iclr_2017 | null | We introduce the hierarchical compositional network (HCN), a directed generative model able to discover and disentangle, without supervision, the building blocks of a set of binary images. The building blocks are binary features defined hierarchically as a composition of some of the features in the layer immediately be... | 2 | {"name": "760.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Miguel Lázaro-Gredilla", "Yi Liu"], "emails": ["miguel@vicarious.com", "yiliu@vicarious.com", "scott@vicarious.com", "dileep@vicarious.com"], "sections": [{"heading": null, "text": "We introduce the hierarchical compositional network (HCN), a... | ||
761 | iclr_2017 | null | Static analyzers are meta-programs that analyze programs to detect potential errors or collect information. For example, they are used as security tools to detect potential buffer overflows. Also, they are used by compilers to verify that a program is well-formed and collect information to generate better code. In this... | 1 | {"name": "761.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["TOY LANGUAGE", "Manzil Zaheer"], "emails": ["manzil.zaheer@cmu.edu", "jean.baptiste.tristan@oracle.com"], "sections": [{"heading": "1 INTRODUCTION", "text": "Can programming language tools, such as static analyzers, be learned from data using... | Not supported with pagination yet | |
762 | iclr_2017 | PERCEIVING FROM LOW FIDELITY VISUAL INPUT | Humans perceive their surroundings in great detail even though most of our visual field is reduced to low-fidelity color-deprived (e.g., dichromatic) input by the retina. In contrast, most deep learning architectures deploy computational resources homogeneously to every part of the visual input. Is such a prodigal depl... | 1 | {"name": "762.pdf", "metadata": {"source": "CRF", "title": "PERCEIVING FROM LOW FIDELITY VISUAL INPUT", "authors": ["Farahnaz A. Wick", "Michael L. Wick"], "emails": ["fwick@cs.umb.edu", "mwick@cs.umass.edu", "mpomplun@cs.umb.edu"], "sections": [{"heading": "1 INTRODUCTION", "text": "Most deep learning architectures pr... | Not supported with pagination yet | |
763 | iclr_2017 | null | We address the problem of modeling multiple simultaneous time series where the observations are correlated not only inside each series, but among the different series. This problem happens in many domains such as ecology, meteorology, etc. We propose a new dynamical state space model, based on representation learning, ... | 1 | {"name": "763.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["SIAN EMBEDDINGS", "Ludovic Dos Santos", "Ludovic Denoyer", "Benjamin Piwowarski", "Patrick Gallinari", "Ali Ziat"], "emails": ["firstname.lastname@lip6.fr", "ali.ziat@vedecom.fr"], "sections": [{"heading": "1 INTRODUCTION", "text": "Relationa... | Not supported with pagination yet | |
765 | iclr_2017 | null | Reinforcement learning is concerned with learning to interact with environments that are initially unknown. State-of-the-art reinforcement learning approaches, such as DQN, are model-free and learn to act effectively across a wide range of environments such as Atari games, but require huge amounts of data. Modelbased t... | 2 | {"name": "765.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["ATARI GAMES", "Felix Leibfried"], "emails": ["felix.leibfried@gmail.com", "nkushman@microsoft.com", "katja.hofmann@microsoft.com"], "sections": [{"heading": "1 INTRODUCTION", "text": "When humans or animals receive reward for taking a particu... | ||
766 | iclr_2017 | null | Most of the pedestrian detection methods are based on hand-crafted features which produce low accuracy on complex scenes. With the development of deep learning method, pedestrian detection has achieved great success. In this paper, we take advantage of a convolutional neural network which is based on Fast R-CNN framewo... | 2 | {"name": "766.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["BATCH NORMALIZATION", "Zhong-Qiu Zhao", "Haiman Bian", "Donghui Hu", "Hervé Glotin"], "emails": ["z.zhao@hfut.edu.cn", "hudh@hfut.edu.cn", "bhm2164@163.com", "h.glotin@gmail.com"], "sections": [{"heading": "1 INTRODUCTION", "text": "In recent... | ||
767 | iclr_2017 | AN ACTOR-CRITIC ALGORITHM FOR LEARNING RATE LEARNING | Stochastic gradient descent (SGD), which updates the model parameters by adding a local gradient times a learning rate at each step, is widely used in model training of machine learning algorithms such as neural networks. It is observed that the models trained by SGD are sensitive to learning rates and good learning ra... | 0 | {"name": "767.pdf", "metadata": {"source": "CRF", "title": "AN ACTOR-CRITIC ALGORITHM FOR LEARNING RATE LEARNING", "authors": ["Chang Xu", "Tao Qin"], "emails": ["changxu@nbjl.nankai.edu.cn", "taoqin@microsoft.com", "wgzwp@nbjl.nankai.edu.cn", "tie-yan.liu@microsoft.com"], "sections": [{"heading": "1 INTRODUCTION", "te... | Not supported with pagination yet | Not supported with pagination yet |
768 | iclr_2017 | TRAINING GROUP ORTHOGONAL NEURAL NETWORKS WITH PRIVILEGED INFORMATION | Learning rich and diverse feature representation are always desired for deep convolutional neural networks (CNNs). Besides, when auxiliary annotations are available for specific data, simply ignoring them would be a great waste. In this paper, we incorporate these auxiliary annotations as privileged information and pro... | 2 | {"name": "768.pdf", "metadata": {"source": "CRF", "title": "TRAINING GROUP ORTHOGONAL NEURAL NETWORKS WITH PRIVILEGED INFORMATION", "authors": ["Yunpeng Chen", "Xiaojie Jin", "Jiashi Feng", "Shuicheng Yan"], "emails": ["chenyunpeng@u.nus.edu", "xiaojie.jin@u.nus.edu", "elefjia@nus.edu.sg", "yanshuicheng@360.cn"], "sect... | ||
770 | iclr_2017 | A NEURAL KNOWLEDGE LANGUAGE MODEL | Current language models have significant limitations in their ability to encode and decode factual knowledge. This is mainly because they acquire such knowledge based on statistical co-occurrences, even if most of the knowledge words are rarely observed named entities. In this paper, we propose a Neural Knowledge Langu... | 1 | {"name": "770.pdf", "metadata": {"source": "CRF", "title": "A NEURAL KNOWLEDGE LANGUAGE MODEL", "authors": ["Sungjin Ahn", "Heeyoul Choi", "Tanel Pärnamaa", "Yoshua Bengio"], "emails": ["heeyoul@gmail.com", "tanel.parnamaa@gmail.com", "yoshua.bengio@umontreal.ca"], "sections": [{"heading": "1 INTRODUCTION", "text": "Ka... | Not supported with pagination yet | |
771 | iclr_2017 | TATIONS WITH A LEXICON | A synonym of a polysemous word is usually only the paraphrase of one sense among many. When lexicons are used to improve vector-space word representations, such paraphrases are unreliable and bring noise to the vector-space. The prior works use a coefficient to adjust the overall learning of the lexicons. They regard t... | 2 | {"name": "771.pdf", "metadata": {"source": "CRF", "title": "TATIONS WITH A LEXICON", "authors": ["Yuanzhi Ke", "Masafumi Hagiwara"], "emails": ["hagiwara}@keio.jp"], "sections": [{"heading": "1 INTRODUCTION", "text": "Vector-space representations of words are reported useful and improve the performance of the machine l... | ||
772 | iclr_2017 | null | Previous work has shown that feature maps of deep convolutional neural networks (CNNs) can be interpreted as feature representation of a particular image region. Features aggregated from these feature maps have been exploited for image retrieval tasks and achieved state-of-the-art performances in recent years. The key ... | 1 | {"name": "772.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["INSTANCE RETRIEVAL", "Jiedong Hao", "Jing Dong", "Wei Wang", "Tieniu Tan"], "emails": [], "sections": [{"heading": null, "text": "Previous work has shown that feature maps of deep convolutional neural networks (CNNs) can be interpreted as fea... | Not supported with pagination yet | |
773 | iclr_2017 | RECTIFIED FACTOR NETWORKS FOR BICLUSTERING | Biclustering is evolving into one of the major tools for analyzing large datasets given as matrix of samples times features. Biclustering has several noteworthy applications and has been successfully applied in life sciences and e-commerce for drug design and recommender systems, respectively. FABIA is one of the most ... | 0 | {"name": "773.pdf", "metadata": {"source": "CRF", "title": "RECTIFIED FACTOR NETWORKS FOR BICLUSTERING", "authors": ["Djork-Arné Clevert", "Thomas Unterthiner"], "emails": ["okko@bioinf.jku.at", "unterthiner@bioinf.jku.at", "hochreit@bioinf.jku.at"], "sections": [{"heading": "1 INTRODUCTION", "text": "Biclustering is w... | Not supported with pagination yet | Not supported with pagination yet |
774 | iclr_2017 | null | We introduce several techniques for sampling and visualizing the latent spaces of generative models. Replacing linear interpolation with spherical linear interpolation prevents diverging from a model’s prior distribution and produces sharper samples. J-Diagrams and MINE grids are introduced as visualizations of manifol... | 1 | {"name": "774.pdf", "metadata": {"source": "CRF", "title": null, "authors": [], "emails": ["tom.white@vuw.ac.nz"], "sections": [{"heading": "1 INTRODUCTION", "text": "Generative models are a popular approach to unsupervised machine learning. Generative neural network models are trained to produce data samples that rese... | Not supported with pagination yet | |
775 | iclr_2017 | LEARNING LOCOMOTION SKILLS USING DEEPRL: DOES | The use of deep reinforcement learning allows for high-dimensional state descriptors, but little is known about how the choice of action representation impacts the learning difficulty and the resulting performance. We compare the impact of four different action parameterizations (torques, muscle-activations, target joi... | 0 | {"name": "775.pdf", "metadata": {"source": "CRF", "title": "LEARNING LOCOMOTION SKILLS USING DEEPRL: DOES", "authors": ["SPACE MATTER", "Xue Bin Peng", "Michiel van de Panne"], "emails": ["xbpeng@cs.ubc.ca", "van@cs.ubc.ca"], "sections": [{"heading": null, "text": "The use of deep reinforcement learning allows for high... | Not supported with pagination yet | Not supported with pagination yet |
776 | iclr_2017 | null | Existing machine translation decoding algorithms generate translations in a strictly monotonic fashion and never revisit previous decisions. As a result, earlier mistakes cannot be corrected at a later stage. In this paper, we present a translation scheme that starts from an initial guess and then makes iterative impro... | 1 | {"name": "776.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["MACHINE TRANSLATION", "Roman Novak", "Michael Auli", "David Grangier"], "emails": [], "sections": [{"heading": "1 INTRODUCTION", "text": "Existing decoding schemes for translation generate outputs either left-to-right, such as for phrasebased... | Not supported with pagination yet | |
777 | iclr_2017 | null | Latent representation learned from multi-layered neural networks via hierarchical feature abstraction enables recent success of deep learning. Under the deep learning framework, generalization performance highly depends on the learned latent representation. In this work, we propose a novel latent space modeling method ... | 1 | {"name": "777.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Hyo-Eun Kim", "Sangheum Hwang"], "emails": ["shwang}@lunit.io", "kyunghyun.cho@nyu.edu"], "sections": [{"heading": "1 INTRODUCTION", "text": "Enhancing the generalization performance against unseen data given some sample data is the main obje... | Not supported with pagination yet | |
778 | iclr_2017 | null | This paper aims to reduce test-time computational load of a deep neural network. Unlike previous methods which factorize a weight matrix into multiple real-valued matrices, our method factorizes both weights and activations into integer and noninteger components. In our method, the real-valued weight matrix is approxim... | 0 | {"name": "778.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Mitsuru Ambai", "Takuya Matsumoto", "Takayoshi Yamashita", "Hironobu Fujiyoshi"], "emails": ["manbai@d-itlab.co.jp", "tmatsumoto@d-itlab.co.jp", "yamashita@cs.chubu.ac.jp", "hf@cs.chubu.ac.jp"], "sections": [{"heading": null, "text": "1 INTRO... | Not supported with pagination yet | Not supported with pagination yet |
779 | iclr_2017 | VOCABULARY SELECTION STRATEGIES FOR NEURAL MACHINE TRANSLATION | Classical translation models constrain the space of possible outputs by selecting a subset of translation rules based on the input sentence. Recent work on improving the efficiency of neural translation models adopted a similar strategy by restricting the output vocabulary to a subset of likely candidates given the sou... | 1 | {"name": "779.pdf", "metadata": {"source": "CRF", "title": "VOCABULARY SELECTION STRATEGIES FOR NEURAL MACHINE TRANSLATION", "authors": ["Gurvan L’Hostis", "David Grangier"], "emails": [], "sections": [{"heading": "1 INTRODUCTION", "text": "Neural Machine Translation (NMT) has made great progress in recent years and im... | Not supported with pagination yet | |
780 | iclr_2017 | null | We consider the problem of how to reduce the cost of communication that is required for the parallel training of a neural network. The state-of-the-art method, Bulk Synchronous Parallel Stochastic Gradient Descent (BSP-SGD), requires many collective communication operations, like broadcasts of parameters or reductions ... | 2 | {"name": "780.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Linnan Wang", "Wei Wu", "Zenglin Xu"], "emails": ["linnan.wang@gatech.edu", "bosilca}@icl.utk.edu", "richie@cc.gatech.edu", "zlxu@uestc.edu.cn"], "sections": [{"heading": "1 INTRODUCTION", "text": "Scaling up neural networks with respect to p... | ||
781 | iclr_2017 | DYNAMIC PARTITION MODELS | We present a new approach for learning compact and intuitive distributed representations with binary encoding. Rather than summing up expert votes as in products of experts, we employ for each variable the opinion of the most reliable expert. Data points are hence explained through a partitioning of the variables into ... | 1 | {"name": "781.pdf", "metadata": {"source": "CRF", "title": "DYNAMIC PARTITION MODELS", "authors": ["Marc Goessling", "Yali Amit"], "emails": ["goessling@galton.uchicago.edu", "amit@galton.uchicago.edu"], "sections": [{"heading": "1 INTRODUCTION", "text": "We consider the task of learning a compact binary representation... | Not supported with pagination yet | |
782 | iclr_2017 | null | In this paper, we propose and investigate a novel memory architecture for neural networks called Hierarchical Attentive Memory (HAM). It is based on a binary tree with leaves corresponding to memory cells. This allows HAM to perform memory access in Θ(log n) complexity, which is a significant improvement over the stand... | 1 | {"name": "782.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["ATTENTIVE MEMORY", "Marcin Andrychowicz", "Google Deepmind", "Karol Kurach"], "emails": [], "sections": [{"heading": "1 INTRO", "text": "Deep Recurrent Neural Networks (RNNs) have recently proven to be very successful in real-word tasks, e.g.... | Not supported with pagination yet | |
783 | iclr_2017 | REVISITING DISTRIBUTED SYNCHRONOUS SGD | Distributed training of deep learning models on large-scale training data is typically conducted with asynchronous stochastic optimization to maximize the rate of updates, at the cost of additional noise introduced from asynchrony. In contrast, the synchronous approach is often thought to be impractical due to idle tim... | 2 | {"name": "783.pdf", "metadata": {"source": "CRF", "title": "REVISITING DISTRIBUTED SYNCHRONOUS SGD", "authors": ["Jianmin Chen", "Xinghao Pan", "Rajat Monga", "Samy Bengio", "Rafal Jozefowicz"], "emails": ["jmchen@google.com", "xinghao@google.com", "rajatmonga@google.com", "bengio@google.com", "rafal@openai.com", "xing... | ||
785 | iclr_2017 | null | We propose a learning method to quantify human intention. Generally, a human being will imagine several potential actions for a given scene, but only one of these actions will subsequently be taken. This makes it difficult to quantify human intentions. To solve this problem, we apply competitive learning to human behav... | 2 | {"name": "785.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Masayoshi Ishikawa", "Mariko Okude", "Takehisa Nishida", "Kazuo Muto"], "emails": ["mariko.okude.uh}@hitachi.com", "kazuo.muto.ny}@hitachi.com"], "sections": [{"heading": null, "text": "We propose a learning method to quantify human intention... | ||
787 | iclr_2017 | null | Multi-label learning aims to automatically assign to an instance (e.g., an image or a document) the most relevant subset of labels from a large set of possible labels. The main challenge is to maintain accurate predictions while scaling efficiently on data sets with extremely large label sets and many training data poi... | 1 | {"name": "787.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Liping Jing", "MiaoMiao Cheng", "Michael W. Mahoney"], "emails": ["lpjing@bjtu.edu.cn", "15112085@bjtu.edu.cn", "11112191@bjtu.edu.cn", "gittens@icsi.berkeley.edu,", "mmahoney@stat.berkeley.edu"], "sections": [{"heading": "1 INTRODUCTION", "t... | Not supported with pagination yet | |
789 | iclr_2017 | null | We focus on generative autoencoders, such as variational or adversarial autoencoders, which jointly learn a generative model alongside an inference model. Generative autoencoders are those which are trained to softly enforce a prior on the latent distribution learned by the inference model. We call the distribution to ... | 2 | {"name": "789.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["MARKOV CHAINS", "Antonia Creswell", "Kai Arulkumaran", "Anil A. Bharath"], "emails": ["ac2211@ic.ac.uk", "ka709@ic.ac.uk", "aab01@ic.ac.uk"], "sections": [{"heading": "1 INTRODUCTION", "text": "Unsupervised learning has benefited greatly from... | ||
790 | iclr_2017 | null | We ask whether neural networks can learn to use secret keys to protect information from other neural networks. Specifically, we focus on ensuring confidentiality properties in a multiagent system, and we specify those properties in terms of an adversary. Thus, a system may consist of neural networks named Alice and Bob... | 2 | {"name": "790.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["David G. Andersen"], "emails": [], "sections": [{"heading": "1 INTRODUCTION", "text": "As neural networks are applied to increasingly complex tasks, they are often trained to meet endto-end objectives that go beyond simple functional specific... | ||
791 | iclr_2017 | DEEP UNSUPERVISED LEARNING THROUGH SPATIAL CONTRASTING | Convolutional networks have marked their place over the last few years as the best performing model for various visual tasks. They are, however, most suited for supervised learning from large amounts of labeled data. Previous attempts have been made to use unlabeled data to improve model performance by applying unsuper... | 0 | {"name": "791.pdf", "metadata": {"source": "CRF", "title": "DEEP UNSUPERVISED LEARNING THROUGH SPATIAL CONTRASTING", "authors": ["Elad Hoffer", "Itay Hubara"], "emails": ["ehoffer@tx.technion.ac.il", "itayh@tx.technion.ac.il", "nailon@cs.technion.ac.il"], "sections": [{"heading": "1 INTRODUCTION", "text": "For the past... | Not supported with pagination yet | Not supported with pagination yet |
792 | iclr_2017 | SOFTTARGET REGULARIZATION AN EFFECTIVE TECHNIQUE TO REDUCE OVER-FITTING IN NEURAL NETWORKS | Deep neural networks are learning models with a very high capacity and therefore prone to over-fitting. Many regularization techniques such as Dropout, DropConnect, and weight decay all attempt to solve the problem of over-fitting by reducing the capacity of their respective models (Srivastava et al., 2014), (Wan et al... | 1 | {"name": "792.pdf", "metadata": {"source": "CRF", "title": "SOFTTARGET REGULARIZATION AN EFFECTIVE TECHNIQUE TO REDUCE OVER-FITTING IN NEURAL NETWORKS", "authors": ["Armen Aghajanyan"], "emails": ["armen.aghajanyan@dimensionalmechanics.com"], "sections": [{"heading": null, "text": "Deep neural networks are learning mod... | Not supported with pagination yet | |
793 | iclr_2017 | null | Recurrent neural nets are widely used for predicting temporal data. Their inherent deep feedforward structure allows learning complex sequential patterns. It is believed that top-down feedback might be an important missing ingredient which in theory could help disambiguate similar patterns depending on broader context.... | 2 | {"name": "793.pdf", "metadata": {"source": "CRF", "title": null, "authors": ["Kamil Rocki"], "emails": ["kmrocki@us.ibm.com"], "sections": [{"heading": "1 INTRODUCTION", "text": "Based on human performance on the same task, it is believed that an important ingredient which is missing in state-of-the-art variants of rec... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.