Program Details‎ > ‎

Conference Program

Listed below are the conference and workshop papers accepted to the International Conference on Learning Representations (ICLR) 2014.

Monday April 14:

08:00 - 09:00 Breakfast - Hawthorne Ballroom

09:00 - 12:30 Oral session - Wildrose Ballroom B&C

09:00 - 09:40 Myths of Representation Learning (invited talk)

    Rich Sutton

09:40 - 10:00 Multilingual Distributed Representations without Word Alignment

    Karl Moritz Hermann; Phil Blunsom

10:00 - 10:20 Zero-Shot Learning by Convex Combination of Semantic Embeddings

    Mohammad Norouzi; Tomas Mikolov; Samy Bengio; Yoram Singer; Jonathon Shlens; Andrea Frome; 

    Greg S. Corrado; Jeffrey Dean

10:20 - 10:50 Break

10:50 - 11:30 Speech Representations: Knowledge or Data? (invited talk)

    Hynek Hermansky

11:30 - 11:50 Exact solutions to the nonlinear dynamics of learning in deep linear neural networks

    Andrew M. Saxe; James L. McClelland; Surya Ganguli

11:50 - 12:10 Revisiting Natural Gradient for Deep Networks

    Razvan Pascanu; Yoshua Bengio

12:10 - 12:30
Unit Tests for Stochastic Optimization
                            Tom Schaul; Ioannis Antonoglou; David Silver 

12:30 - 17:30 Lunch on own / free time

17:30 - 19:00 Dinner - Hawthorne Ballroom

19:00 - 22:00 Poster session I - Wildrose Ballroom A

Conference Posters: [Note for presenters: Poster board size: 4' x 8']

The return of AdaBoost.MH: multi-class Hamming trees

    Balázs Kégl

Neuronal Synchrony in Complex-Valued Deep Networks

    David P. Reichert; Thomas Serre

Bounding the Test Log-Likelihood of Generative Models

    Yoshua Bengio; Li Yao; KyungHyun Cho

A Generative Product-of-Filters Model of Audio

    Dawen Liang; Mathew D. Hoffman; Gautham Mysore

How to Construct Deep Recurrent Neural Networks

    Razvan Pascanu; Caglar Gulcehre; Kyunghyun Cho; Yoshua Bengio

Zero-Shot Learning and Clustering for Semantic Utterance Classification

    Yann N. Dauphin; Gokhan Tur; Dilek Hakkani-Tur; Larry Heck

An empirical analysis of dropout in piecewise linear networks

    David Warde-Farley; Ian J. Goodfellow; Aaron Courville; Yoshua Bengio

An Empirical Investigation of Catastrophic Forgeting in Gradient-Based Neural Networks

    Ian J. Goodfellow; Mehdi Mirza; Da Xiao; Aaron Courville; Yoshua Bengio;

On Fast Dropout and its Applicability to Recurrent Networks

    Justin Bayer; Christian Osendorfer; Daniela Korhammer; Nutan Chen; Sebastian Urban; Patrick van der Smagt

Network In Network

    Min Lin; Qiang Chen; Shuicheng Yan

Workshop Posters: [Note for presenters: Poster board size: 4' x 8']

Deep Learning Embeddings for Discontinuous Linguistic Units

    Wenpeng Yin; Hinrich Schütze

Learning Factored Representations in a Deep Mixture of Experts

    David Eigen; Marc'Aurelio Ranzato; Ilya Sutskever

Learning Information Spread in Content Networks

    Cédric Lagnier; Simon Bourigault; Sylvain Lamprier; Ludovic Denoyer; Patrick Gallinari

Learning States Representations in POMDP

    Gabriella Contardo; Ludovic Denoyer; Thierry Artieres; Patrick Gallinari

Distributional Models and Deep Learning Embeddings: Combining the Best of Both Worlds

    Irina Sergienya; Hinrich Schütze

Stochastic Gradient Estimate Variance in Contrastive Divergence and Persistent Contrastive Divergence

    Mathias Berglund; Tapani Raiko

Stopping Criteria in Contrastive Divergence: Alternatives to the Reconstruction Error

    David Buchaca; Enrique Romero; Ferran Mazzanti; Jordi Delgado

Continuous Learning: Engineering Super Features With Feature Algebras

    Michael Tetelman

Multimodal Transitions for Generative Stochastic Networks

    Sherjil Ozair; Li Yao; Yoshua Bengio

Factorial Hidden Markov Models for Learning Representations of Natural Language

    Anjan Nepal; Alexander Yates

Can recursive neural tensor networks learn logical reasoning?

    Samuel R. Bowman

A Primal-Dual Method for Training Recurrent Neural Networks Constrained by the Echo-State Property

    Jianshu Chen; Li Deng

Principled Non-Linear Feature Selection

    Dimitrios Athanasakis; John Shawe-Taylor; Delmiro Fernandez-Reyes

Rate-Distortion Auto-Encoders

    Luis G. Sanchez Giraldo; Jose C. Principe

Low-Rank Approximations for Conditional Feedforward Computation in Deep Neural Networks

    Andrew Davis; Itamar Arel

Semistochastic Quadratic Bound Methods

    Aleksandr Y. Aravkin; Anna Choromanska; Tony Jebara; Dimitri Kanevsky

An Architecture for Distinguishing between Predictors and Inhibitors in Reinforcement Learning

    Patrick C. Connor; Thomas P. Trappenberg

Tuesday April 15:

08:00 - 09:00 Breakfast - Hawthorne Ballroom

09:00 - 12:40 Oral session- Wildrose B&C

09:00 - 09:40 Symmetry-Based Learning  (invited talk)

    Pedro Domingos

09:40 - 10:00 Auto-Encoding Variational Bayes

    Diederik P. Kingma; Max Welling

10:00 - 10:20 Group-sparse Embeddings in Collective Matrix Factorization

    Arto Klami; Guillaume Bouchard; Abhishek Tripathi

10:20 - 10:50 Break

10:50 - 11:30 Learning Visual Representations at Scale  (invited talk)

    Vincent Vanhoucke

11:30 - 11:45 Relaxations for inference in restricted Boltzmann machines

    Sida I. Wang; Roy Frostig; Percy Liang; Christopher D. Manning

11:45 - 12:00 Learning Semantic Script Knowledge with Event Embeddings

    Ashutosh Modi; Ivan Titov

12:00 - 12:15 Unsupervised Feature Learning by Deep Sparse Coding

    Yunlong He; Koray Kavukcuoglu; Yun Wang; Arthur Szlam; Yanjun Qi

12:15 - 12:30 End-to-End Text Recognition with Hybrid HMM Maxout Models

    Ouais Alsharif; Joelle Pineau

12:40 - 17:30 Lunch on own / free time

17:30 - 19:00 Dinner - Hawthorne Ballroom

19:00 - 22:00 Poster session II - Wildrose Ballroom A

Conference Posters: [Note for presenters: Poster board size: 4' x 8']

Learning Human Pose Estimation Features with Convolutional Networks

    Ajrun Jain; Jonathan Tompson; Mykhaylo Andriluka; Graham W. Taylor; Christoph Bregler

EXMOVES: Classifier-based Features for Scalable Action Recognition

    Du Tran; Lorenzo Torresani

On the number of inference regions of deep feed forward networks with piece-wise linear activations

    Razvan Pascanu; Guido Montufar; Yoshua Bengio

Intriguing properties of neural networks

    Christian Szegedy; Wojciech Zaremba; Ilya Sutskever; Joan Bruna; Dumitru Erhan; Ian Goodfellow; Rob Fergus

Fast Training of Convolutional Networks through FFTs

    Michael Mathieu; Mikael Henaff; Yann LeCun

Deep and Wide Multiscale Recursive Networks for Robust Image Labeling

    Gary B. Huang; Viren Jain

Some Improvements on Deep Convolutional Neural Network Based Image Classification

    Andrew G. Howard

Deep Convolutional Ranking for Multilabel Image Annotation

    Yunchao Gong; Yangqing Jia; Thomas Leung; Alexander Toshev; Sergey Ioffe

Learning to encode motion using spatio-temporal synchrony

    Kishore Reddy Konda; Roland Memisevic; Vincent Michalski

Multi-View Priors for Learning Detectors from Sparse Viewpoint Data

    Bojan Pepik; Michael Stark; Peter Gehler; Bernt Schiele

k-Sparse Autoencoders

    Alireza Makhzani; Brendan Frey

Workshop Posters: [Note for presenters: Poster board size: 4' x 8']

Multi-GPU Training of ConvNets

    Omry Yadan; Keith Adams; Yaniv Taigman; Marc'Aurelio Ranzato

GPU Asynchronous Stochastic Gradient Descent to Speed Up Neural Network Training

    Thomas Huang; Zhe Lin; Hailin Jin; Jianchao Yang; Thomas Paine

Generic Deep Networks with Wavelet Scattering

    Edouard Oyallon; Stéphane Mallat; Laurent Sifre

Deep learning for class-generic object detection

    Brody Huval; Adam Coates; Andrew Ng

Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

    Karen Simonyan; Andrea Vedaldi; Andrew Zisserman

Unsupervised feature learning by augmenting single images

    Alexey Dosovitskiy; Jost Tobias Springenberg; Thomas Brox

Correlation-based construction of neighborhood and edge features

    Balázs Kégl

Modeling correlations in spontaneous activity of visual cortex with centered Gaussian-binary deep Boltzmann machines

    Nan Wang; Laurenz Wiskott; Dirk Jancke

Understanding Deep Architectures using a Recursive Convolutional Network

    David Eigen; Jason Rolfe; Rob Fergus; Yann LeCun

Learning High-level Image Representation for Image Retrieval via Multi-Task DNN using Clickthrough Data

    Wei Yu; Tiejun Zhao; Yalong Bai; Wei-Ying Ma; Kuiyuan Yang

One-Shot Adaptation of Supervised Deep Convolutional Models

    Judy Hoffman; Eric Tzeng; Jeff Donahue; Yangqing Jia; Kate Saenko; Trevor Darrell

Improving Deep Neural Networks with Probabilistic Maxout Units

    Jost Tobias Springenberg; Martin Riedmiller

Efficient Visual Coding: From Retina To V2

    Honghao Shan; Garrison Cottrell

Deep learning for neuroimaging: a validation study

    Sergey M. Plis; Devon R. Hjelm; Ruslan Salakhutdinov; Vince D. Calhoun

Image Representation Learning Using Graph Regularized Auto-Encoders

    Yiyi Liao; Yue Wang; Yong Liu

Deep Belief Networks for Image Denoising

    Mohammad Ali Keyvanrad; Mohammad Pezeshki; Mohammad Mehdi Homayounpour

Approximated Infomax Early Stopping: Revisiting Gaussian RBMs on Natural Images

    Taichi Kiwaki; Takaki Makino; Kazuyuki Aihara

Wednesday April 16:

08:00 - 09:00 Breakfast - Hawthorne Ballroom

09:00 - 12:30 Oral session - Wildrose Ballroom B&C

09:00 - 09:20 OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks

Pierre Sermanet; Rob Fergus; Yann LeCun; Xiang Zhang; David Eigen; Michael Mathieu

09:20 - 09:40 Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks

Ian J. Goodfellow; Yaroslav Bulatov; Julian Ibarz; Sacha Arnoud; Vinay Shet

09:40 - 10:00 Sequentially Generated Instance-Dependent Image Representations for Classification

Ludovic Denoyer; Matthieu Cord; Patrick Gallinari; Nicolas Thome; Gabriel Dulac-Arnold

10:00 - 10:20 Learned versus Hand-Designed Feature Representations for 3d Agglomeration

John A. Bogovic; Gary B. Huang; Viren Jain

10:20 - 10:50 Break

10:50 - 11:30 Representing Relations (invited talk)                         Roland Memisevic

11:30 - 11:50 Spectral Networks and Locally Connected Networks on Graphs

Joan Bruna; Wojciech Zaremba; Arthur Szlam; Yann LeCun

11:50 - 12:10 Sparse similarity-preserving hashing

Alex M. Bronstein; Pablo Sprechmann; Michael M. Bronstein; Jonathan Masci; Guillermo Sapiro

12:10 - 12:30 Learning Transformations for Classification Forests

Qiang Qiu; Guillermo Sapiro

12:30 - 17:30 Lunch on own / free time

17:30 - 19:20 Banquet - Hawthorne Ballroom

19:30 - 20:10 (Town-Hall meeting) - Hawthorne Ballroom