ICLR 2017

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

iclr2017:workshop_posters [2017/03/27 17:41]
hugo
iclr2017:workshop_posters [2017/04/23 09:27] (current)
hugo
Line 1: Line 1:
-=====Workshop Poster Sessions=====+======Workshop Poster Sessions======
  
 Below are the Workshop Track papers presented at each of the poster sessions (on Monday, Tuesday or Wednesday, in the morning or evening). To find a paper, look for the poster with the corresponding number in the area dedicated to the Workshop Track. Below are the Workshop Track papers presented at each of the poster sessions (on Monday, Tuesday or Wednesday, in the morning or evening). To find a paper, look for the poster with the corresponding number in the area dedicated to the Workshop Track.
  
-===Monday Morning (April 24th, 10:​30am ​to 12:30pm)=== +======Note ​to the Presenters======= 
-  - Extrapolation ​and learning equations +Each poster panel is 2 meters large and 1 meter tall.\\ 
-  - Effectiveness of Transfer Learning in EHR data +If needed, tape will be provided ​to fix your poster.
-  - Intelligent synapses for multi-task and transfer learning +
-  - Unsupervised and Efficient Neural Graph Model with Distributed Representations +
-  - Accelerating SGD for Distributed Deep-Learning Using an Approximted Hessian Matrix +
-  - Accelerating Eulerian Fluid Simulation With Convolutional Networks +
-  - Forced ​to Learn: Discovering Disentangled Representations Without Exhaustive Labels +
-  - Deep Nets Don't Learn via Memorization +
-  - Learning Algorithms for Active Learning +
-  - Reinterpreting Importance-Weighted Autoencoders +
-  - Robustness to Adversarial Examples through an Ensemble of Specialists +
-  - Neural Expectation Maximization +
-  - On Hyperparameter Optimization in Learning Systems +
-  - Recurrent Normalization Propagation +
-  - Joint Training of Ratings and Reviews with Recurrent Recommender Networks +
-  - Towards an Automatic Turing Test: Learning to Evaluate Dialogue Responses +
-  - Joint Embeddings of Scene Graphs and Images +
-  - Unseen Style Transfer Based on a Conditional Fast Style Transfer Network+
  
-===Monday ​Afternoon ​(April 24th, 4:30pm to 6:30pm)=== +<​html><​div id='​monday_morning'></​div></​html>​ 
-  Audio Super-Resolution using Neural ​Networks +====Monday ​Morning ​(April 24th, 10:30am to 12:30pm)==== 
-  - Semantic embeddings ​for program behaviour patterns +W1: Extrapolation and learning equations\\ 
-  ​De novo drug design with deep generative models : an empirical study +W2: Effectiveness of Transfer Learning in EHR data\\ 
-  - Memory Matching ​Networks ​for Genomic Sequence Classification +W3: Intelligent synapses for multi-task and transfer learning\\ 
-  - Char2WavEnd-to-End Speech Synthesis +W4: Unsupervised and Efficient ​Neural ​Graph Model with Distributed Representations\\ 
-  - Fast Chirplet Transform Injects Priors ​in Deep Learning ​of Animal Calls and Speech +W5: Accelerating SGD for Distributed Deep-Learning Using an Approximted Hessian Matrix\\ 
-  Weight-averaged consistency targets improve semi-supervised deep learning results +W6: Accelerating Eulerian Fluid Simulation With Convolutional ​Networks\\ 
-  - Particle Value Functions +W7Forced ​to Learn: Discovering Disentangled Representations Without Exhaustive Labels\\ 
-  - Out-of-class novelty generation: an experimental foundation +W8: Dataset Augmentation ​in Feature Space\\ 
-  - Performance guarantees for transferring representations +W9: Learning ​Algorithms for Active Learning\\ 
-  - Generative Adversarial ​Learning ​of Markov Chains +W10: Reinterpreting Importance-Weighted Autoencoders\\ 
-  - Short and DeepSketching and Neural Networks +W11Robustness to Adversarial Examples through ​an Ensemble of Specialists\\ 
-  - Understanding intermediate layers using linear classifier probes +W12: (empty) \\ 
-  - Symmetry-Breaking Convergence Analysis ​of Certain Two-layered Neural ​Networks ​with ReLU nonlinearity +W13: On Hyperparameter Optimization in Learning ​Systems\\ 
-  - Neural Combinatorial Optimization with Reinforcement ​Learning +W14Recurrent Normalization Propagation\\ 
-  - Tactics ​of Adversarial Attacks ​on Deep Reinforcement Learning Agents +W15: Joint Training ​of Ratings and Reviews with Recurrent Recommender ​Networks\\ 
-  - Adversarial Discriminative Domain Adaptation (workshop extended abstract) +W16: Towards an Automatic Turing Test: Learning ​to Evaluate Dialogue Responses\\ 
-  - Efficient Sparse-Winograd Convolutional Neural Networks+W17: Joint Embeddings ​of Scene Graphs and Images\\ 
 +W18: Unseen Style Transfer Based on a Conditional Fast Style Transfer Network\\
  
-===Tuesday Morning (April 25th, 10:30am to 12:30pm)=== 
-  - Programming With a Differentiable Forth Interpreter 
-  - Unsupervised Feature Learning for Audio Analysis 
-  - Neural Functional Programming 
-  - A Smooth Optimisation Perspective on Training Feedforward Neural Networks 
-  - Synthetic Gradient Methods with Virtual Forward-Backward Networks 
-  - Explaining the Learning Dynamics of Direct Feedback Alignment 
-  - Training a Subsampling Mechanism in Expectation 
-  - Deep Kernel Machines via the Kernel Reparametrization Trick 
-  - Encoding and Decoding Representations with Sum- and Max-Product Networks 
-  - Embracing Data Abundance 
-  - Variational Intrinsic Control 
-  - Fast Adaptation in Generative Models with Generative Matching Networks 
-  - Efficient variational Bayesian neural network ensembles for outlier detection 
-  - Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols 
-  - Adaptive Feature Abstraction for Translating Video to Language 
-  - Delving into adversarial attacks on deep policies 
-  - Tuning Recurrent Neural Networks with Reinforcement Learning 
-  - DeepMask: Masking DNN Models for robustness against adversarial samples 
-  - Restricted Boltzmann Machines provide an accurate metric for retinal responses to visual stimuli 
  
-===Tuesday ​Afternoon (April ​25th, 4:30pm to 6:​30pm)=== +<​html><​div id='​monday_afternoon'></​div></​html>​ 
-  - Lifelong Perceptual Programming By Example +====Monday ​Afternoon (April ​24th, 4:30pm to 6:30pm)==== 
-  - Neu0 +W1: Audio Super-Resolution using Neural ​Networks\\ 
-  - Dance Dance Convolution +W2: Semantic embeddings ​for program behaviour patterns\\ 
-  ​Bit-Pragmatic Deep Neural ​Network Computing +W3: De novo drug design with deep generative models : an empirical study\\ 
-  - On Improving the Numerical Stability of Winograd Convolutions +W4: Memory Matching ​Networks ​for Genomic Sequence Classification\\ 
-  - Fast Generation ​for Convolutional Autoregressive Models +W5: Char2Wav: End-to-End Speech Synthesis\\ 
-  - THE PREIMAGE OF RECTIFIER NETWORK ACTIVITIES +W6: Fast Chirplet Transform Injects Priors ​in Deep Learning ​of Animal Calls and Speech\\ 
-  - Training Triplet ​Networks ​with GAN +W7: Weight-averaged consistency targets improve semi-supervised ​deep learning ​results\\ 
-  On Robust Concepts and Small Neural Nets +W8: Particle Value Functions\\ 
-  ​Pl@ntNet app in the era of deep learning +W9: Out-of-class novelty generation: an experimental foundation\\ 
-  - Exponential Machines +W10: Performance guarantees ​for transferring representations\\ 
-  Online Multi-Task Learning Using Biased Sampling +W11: Generative Adversarial Learning ​of Markov Chains\\ 
-  - Online Structure Learning ​for Sum-Product Networks with Gaussian Leaves +W12: Short and Deep: Sketching and Neural Networks\\ 
-  - A Theoretical Framework for Robustness ​of (Deep) Classifiers against Adversarial Samples +W13: Understanding intermediate layers using linear classifier probes\\ 
-  - Compositional Kernel Machines +W14Symmetry-Breaking Convergence Analysis of Certain Two-layered Neural Networks with ReLU nonlinearity\\ 
-  - Loss is its own RewardSelf-Supervision for Reinforcement Learning +W15: Neural Combinatorial Optimization with Reinforcement Learning\\ 
-  - Changing Model Behavior at Test-time Using Reinforcement Learning +W16: Tactics of Adversarial Attacks on Deep Reinforcement Learning ​Agents\\ 
-  - Precise Recovery of Latent Vectors from Generative ​Adversarial ​Networks +W17: Adversarial ​Discriminative Domain Adaptation (workshop extended abstract)\\ 
-  Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization+W18: Efficient Sparse-Winograd Convolutional Neural Networks\\ 
 +W19: Neural Expectation Maximization\\
  
-===Wednesday Morning (April 26th, 10:30am to 12:30pm)=== 
-  - NEUROGENESIS-INSPIRED DICTIONARY LEARNING: ONLINE MODEL ADAPTION IN A CHANGING WORLD 
-  - The High-Dimensional Geometry of Binary Neural Networks 
-  - Discovering objects and their relations from entangled scene representations 
-  - A Differentiable Physics Engine for Deep Learning in Robotics 
-  - Automated Generation of Multilingual Clusters for the Evaluation of Distributed Representations 
-  - Development of JavaScript-based deep learning platform and application to distributed training 
-  - Factorization tricks for LSTM networks 
-  - Shake-Shake regularization of 3-branch residual networks 
-  - Trace Norm Regularised Deep Multi-Task Learning 
-  - Deep Learning with Sets and Point Clouds 
-  - Dataset Augmentation in Feature Space 
-  - Multiplicative LSTM for sequence modelling 
-  - Learning to Discover Sparse Graphical Models 
-  - Revisiting Batch Normalization For Practical Domain Adaptation 
-  - Early Methods for Detecting Adversarial Images and a Colorful Saliency Map 
-  - Natural Language Generation in Dialogue using Lexicalized and Delexicalized Data 
-  - Coupling Distributed and Symbolic Execution for Natural Language Queries 
-  - Adversarial Examples for Semantic Image Segmentation 
-  - RenderGAN: Generating Realistic Labeled Data 
  
-===Wednesday Afternoon (April 26th, 4:30pm to 6:​30pm)=== +<​html><​div id='​tuesday_morning'></​div></​html>​ 
-  ​- ​Song From PI: A Musically Plausible Network for Pop Music Generation +====Tuesday Morning (April 25th, 10:30am to 12:​30pm)==== 
-  ​- ​Charged Point Normalization:​ An Efficient Solution to the Saddle Point Problem +W1: Programming With a Differentiable Forth Interpreter\\ 
-  ​- ​Towards "​AlphaChem":​ Chemical Synthesis Planning with Tree Search and Deep Neural Network Policies +W2: Unsupervised Feature Learning for Audio Analysis\\ 
-  ​- ​CommAI: Evaluating the first steps towards a useful general AI +W3: Neural Functional Programming\\ 
-  ​- ​Joint Multimodal Learning with Deep Generative Models +W4: A Smooth Optimisation Perspective on Training Feedforward Neural Networks\\ 
-  ​- ​Transferring Knowledge to Smaller Network with Class-Distance Loss +W5: Synthetic Gradient Methods with Virtual Forward-Backward Networks\\ 
-  ​- ​Regularizing Neural Networks by Penalizing Confident Output Distributions +W6: Explaining the Learning Dynamics of Direct Feedback Alignment\\ 
-  ​- ​Adversarial Attacks on Neural Network Policies +W7: Training a Subsampling Mechanism in Expectation\\ 
-  ​- ​Generalizable Features From Unsupervised Learning +W8: Deep Kernel Machines via the Kernel Reparametrization Trick\\ 
-  ​- ​Compact Embedding of Binary-coded Inputs and Outputs using Bloom Filters +W9: Encoding and Decoding Representations with Sum- and Max-Product Networks\\ 
-  ​- ​Semi-supervised deep learning by metric embedding +W10: Embracing Data Abundance\\ 
-  - REBARLow-variance, unbiased gradient estimates for discrete latent variable models +W11: Variational Intrinsic Control\\ 
-  ​- ​Variational Reference Priors +W12: Fast Adaptation in Generative Models with Generative Matching Networks\\ 
-  ​- ​Gated Multimodal Units for Information Fusion +W13: Efficient variational Bayesian neural network ensembles for outlier detection\\ 
-  ​- ​Playing SNES in the Retro Learning Environment +W14: Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols\\ 
-  ​- ​Unsupervised Perceptual Rewards for Imitation Learning +W15: Adaptive Feature Abstraction for Translating Video to Language\\ 
-  ​- ​Perception Updating Networks: On architectural constraints for interpretable video generative models +W16: Delving into adversarial attacks on deep policies\\ 
-  ​- ​Adversarial examples in the physical world+W17: Tuning Recurrent Neural Networks with Reinforcement Learning\\ 
 +W18: DeepMask: Masking DNN Models for robustness against adversarial samples\\ 
 +W19: Restricted Boltzmann Machines provide an accurate metric for retinal responses to visual stimuli\\ 
 + 
 +<​html><​div id='​tuesday_afternoon'></​div></​html>​ 
 +====Tuesday Afternoon (April 25th, 2:00pm to 4:​00pm)==== 
 +W1: Lifelong Perceptual Programming By Example\\ 
 +W2: Neu0\\ 
 +W3: Dance Dance Convolution\\ 
 +W4: Bit-Pragmatic Deep Neural Network Computing\\ 
 +W5: On Improving the Numerical Stability of Winograd Convolutions\\ 
 +W6: Fast Generation for Convolutional Autoregressive Models\\ 
 +W7: THE PREIMAGE OF RECTIFIER NETWORK ACTIVITIES\\ 
 +W8: Training Triplet Networks with GAN\\ 
 +W9: On Robust Concepts and Small Neural Nets\\ 
 +W10: Pl@ntNet app in the era of deep learning\\ 
 +W11: Exponential Machines\\ 
 +W12: Online Multi-Task Learning Using Biased Sampling\\ 
 +W13: Online Structure Learning for Sum-Product Networks with Gaussian Leaves\\ 
 +W14: A Theoretical Framework for Robustness of (Deep) Classifiers against Adversarial Samples\\ 
 +W15: Compositional Kernel Machines\\ 
 +W16: Loss is its own Reward: Self-Supervision for Reinforcement Learning\\ 
 +W17: REBAR: Low-variance,​ unbiased gradient estimates for discrete latent variable models\\ 
 +W18: Precise Recovery of Latent Vectors from Generative Adversarial Networks\\ 
 +W19: Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization\\ 
 + 
 +<​html><​div id='​wednesday_morning'></​div></​html>​ 
 +====Wednesday Morning (April 26th, 10:30am to 12:​30pm)==== 
 +W1: NEUROGENESIS-INSPIRED DICTIONARY LEARNING: ONLINE MODEL ADAPTION IN A CHANGING WORLD\\ 
 +W2: The High-Dimensional Geometry of Binary Neural Networks\\ 
 +W3: Discovering objects and their relations from entangled scene representations\\ 
 +W4: A Differentiable Physics Engine for Deep Learning in Robotics\\ 
 +W5: Automated Generation of Multilingual Clusters for the Evaluation of Distributed Representations\\ 
 +W6: Development of JavaScript-based deep learning platform and application to distributed training\\ 
 +W7: Factorization tricks for LSTM networks\\ 
 +W8: Shake-Shake regularization of 3-branch residual networks\\ 
 +W9: Trace Norm Regularised Deep Multi-Task Learning\\ 
 +W10: Deep Learning with Sets and Point Clouds\\ 
 +W11: Deep Nets Don't Learn via Memorization\\ 
 +W12: Multiplicative LSTM for sequence modelling\\ 
 +W13: Learning to Discover Sparse Graphical Models\\ 
 +W14: Revisiting Batch Normalization For Practical Domain Adaptation\\ 
 +W15: Early Methods for Detecting Adversarial Images and a Colorful Saliency Map\\ 
 +W16: Natural Language Generation in Dialogue using Lexicalized and Delexicalized Data\\ 
 +W17: Coupling Distributed and Symbolic Execution for Natural Language Queries\\ 
 +W18: Adversarial Examples for Semantic Image Segmentation\\ 
 +W19: RenderGAN: Generating Realistic Labeled Data\\ 
 + 
 +<​html><​div id='​wednesday_afternoon'></​div></​html>​ 
 +====Wednesday Afternoon (April 26th, 4:30pm to 6:30pm)==== 
 +W1: Song From PI: A Musically Plausible Network for Pop Music Generation\\ 
 +W2: Charged Point Normalization:​ An Efficient Solution to the Saddle Point Problem\\ 
 +W3: Towards "​AlphaChem":​ Chemical Synthesis Planning with Tree Search and Deep Neural Network Policies\\ 
 +W4: CommAI: Evaluating the first steps towards a useful general AI\\ 
 +W5: Joint Multimodal Learning with Deep Generative Models\\ 
 +W6: Transferring Knowledge to Smaller Network with Class-Distance Loss\\ 
 +W7: Regularizing Neural Networks by Penalizing Confident Output Distributions\\ 
 +W8: Adversarial Attacks on Neural Network Policies\\ 
 +W9: Generalizable Features From Unsupervised Learning\\ 
 +W10: Compact Embedding of Binary-coded Inputs and Outputs using Bloom Filters\\ 
 +W11: Semi-supervised deep learning by metric embedding\\ 
 +W12Changing Model Behavior at Test-time Using Reinforcement Learning\\ 
 +W13: Variational Reference Priors\\ 
 +W14: Gated Multimodal Units for Information Fusion\\ 
 +W15: Playing SNES in the Retro Learning Environment\\ 
 +W16: Unsupervised Perceptual Rewards for Imitation Learning\\ 
 +W17: Perception Updating Networks: On architectural constraints for interpretable video generative models\\ 
 +W18: Adversarial examples in the physical world\\