Search Results for author: Nicolas Ballas

Found 42 papers, 23 papers with code

Proposal-based Video Completion

no code implementations ECCV 2020 Yuan-Ting Hu, Heng Wang, Nicolas Ballas, Kristen Grauman, Alexander G. Schwing

Video inpainting is an important technique for a wide variety of applications from video content editing to video restoration.

Image Inpainting object-detection +4

Predicting masked tokens in stochastic locations improves masked image modeling

no code implementations31 Jul 2023 Amir Bar, Florian Bordes, Assaf Shocher, Mahmoud Assran, Pascal Vincent, Nicolas Ballas, Trevor Darrell, Amir Globerson, Yann Lecun

Specifically, we condition the model on stochastic masked token positions to guide the model toward learning features that are more robust to location uncertainties.

Language Modelling Masked Language Modeling +3

A Simple Recipe for Competitive Low-compute Self supervised Vision Models

no code implementations23 Jan 2023 Quentin Duval, Ishan Misra, Nicolas Ballas

Our main insight is that existing joint-embedding based SSL methods can be repurposed for knowledge distillation from a large self-supervised teacher to a small student model.

Knowledge Distillation

ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotations

no code implementations3 Nov 2022 Badr Youbi Idrissi, Diane Bouchacourt, Randall Balestriero, Ivan Evtimov, Caner Hazirbas, Nicolas Ballas, Pascal Vincent, Michal Drozdzal, David Lopez-Paz, Mark Ibrahim

Equipped with ImageNet-X, we investigate 2, 200 current recognition models and study the types of mistakes as a function of model's (1) architecture, e. g. transformer vs. convolutional, (2) learning paradigm, e. g. supervised vs. self-supervised, and (3) training procedures, e. g., data augmentation.

Data Augmentation

Neural Attentive Circuits

no code implementations14 Oct 2022 Nasim Rahaman, Martin Weiss, Francesco Locatello, Chris Pal, Yoshua Bengio, Bernhard Schölkopf, Li Erran Li, Nicolas Ballas

Recent work has seen the development of general purpose neural architectures that can be trained to perform tasks across diverse data modalities.

Point Cloud Classification text-classification +1

The Hidden Uniform Cluster Prior in Self-Supervised Learning

no code implementations13 Oct 2022 Mahmoud Assran, Randall Balestriero, Quentin Duval, Florian Bordes, Ishan Misra, Piotr Bojanowski, Pascal Vincent, Michael Rabbat, Nicolas Ballas

A successful paradigm in representation learning is to perform self-supervised pretraining using tasks based on mini-batch statistics (e. g., SimCLR, VICReg, SwAV, MSN).

Clustering Representation Learning +1

BARACK: Partially Supervised Group Robustness With Guarantees

no code implementations31 Dec 2021 Nimit S. Sohoni, Maziar Sanjabi, Nicolas Ballas, Aditya Grover, Shaoliang Nie, Hamed Firooz, Christopher Ré

Theoretically, we provide generalization bounds for our approach in terms of the worst-group performance, which scale with respect to both the total number of training points and the number of training points with group labels.

Fairness Generalization Bounds

Trade-offs of Local SGD at Scale: An Empirical Study

no code implementations15 Oct 2021 Jose Javier Gonzalez Ortiz, Jonathan Frankle, Mike Rabbat, Ari Morcos, Nicolas Ballas

As datasets and models become increasingly large, distributed training has become a necessary component to allow deep neural networks to train in reasonable amounts of time.

Image Classification

INFERNO: Inferring Object-Centric 3D Scene Representations without Supervision

no code implementations29 Sep 2021 Lluis Castrejon, Nicolas Ballas, Aaron Courville

Each object representation defines a localized neural radiance field that is used to generate 2D views of the scene through a differentiable rendering process.

Video Object Tracking Visual Reasoning

Hierarchical Video Generation for Complex Data

no code implementations4 Jun 2021 Lluis Castrejon, Nicolas Ballas, Aaron Courville

Inspired by this we propose a hierarchical model for video generation which follows a coarse to fine approach.

Video Generation

SSW-GAN: Scalable Stage-wise Training of Video GANs

no code implementations1 Jan 2021 Lluis Castrejon, Nicolas Ballas, Aaron Courville

Current state-of-the-art generative models for videos have high computational requirements that impede high resolution generations beyond a few frames.

A Closer Look at Codistillation for Distributed Training

no code implementations6 Oct 2020 Shagun Sodhani, Olivier Delalleau, Mahmoud Assran, Koustuv Sinha, Nicolas Ballas, Michael Rabbat

Surprisingly, we find that even at moderate batch sizes, models trained with codistillation can perform as well as models trained with synchronous data-parallel methods, despite using a much weaker synchronization mechanism.

Distributed Computing

Revisiting Loss Modelling for Unstructured Pruning

1 code implementation22 Jun 2020 César Laurent, Camille Ballas, Thomas George, Nicolas Ballas, Pascal Vincent

By removing parameters from deep neural networks, unstructured pruning methods aim at cutting down memory footprint and computational cost, while maintaining prediction accuracy.

Supervision Accelerates Pre-training in Contrastive Semi-Supervised Learning of Visual Representations

2 code implementations18 Jun 2020 Mahmoud Assran, Nicolas Ballas, Lluis Castrejon, Michael Rabbat

We investigate a strategy for improving the efficiency of contrastive learning of visual representations by leveraging a small amount of supervised information during pre-training.

Contrastive Learning

SlowMo: Improving Communication-Efficient Distributed SGD with Slow Momentum

1 code implementation ICLR 2020 Jianyu Wang, Vinayak Tantia, Nicolas Ballas, Michael Rabbat

We provide theoretical convergence guarantees showing that SlowMo converges to a stationary point of smooth non-convex losses.

Blocking Distributed Optimization +3

Needles in Haystacks: On Classifying Tiny Objects in Large Images

1 code implementation16 Aug 2019 Nick Pawlowski, Suvrat Bhooshan, Nicolas Ballas, Francesco Ciompi, Ben Glocker, Michal Drozdzal

In some important computer vision domains, such as medical or hyperspectral imaging, we care about the classification of tiny objects in large images.

Classification General Classification +1

Gossip-based Actor-Learner Architectures for Deep Reinforcement Learning

1 code implementation NeurIPS 2019 Mahmoud Assran, Joshua Romoff, Nicolas Ballas, Joelle Pineau, Michael Rabbat

We show that we can run several loosely coupled GALA agents in parallel on a single GPU and achieve significantly higher hardware utilization and frame-rates than vanilla A2C at comparable power draws.

reinforcement-learning Reinforcement Learning (RL)

Improved Conditional VRNNs for Video Prediction

1 code implementation ICCV 2019 Lluis Castrejon, Nicolas Ballas, Aaron Courville

To address this issue, we propose to increase the expressiveness of the latent distributions and to use higher capacity likelihood models.

Video Generation Video Prediction

Fast Approximate Natural Gradient Descent in a Kronecker Factored Eigenbasis

no code implementations NeurIPS 2018 Thomas George, César Laurent, Xavier Bouthillier, Nicolas Ballas, Pascal Vincent

Optimization algorithms that leverage gradient covariance information, such as variants of natural gradient descent (Amari, 1998), offer the prospect of yielding more effective descent directions.

Stochastic Gradient Push for Distributed Deep Learning

2 code implementations ICLR 2019 Mahmoud Assran, Nicolas Loizou, Nicolas Ballas, Michael Rabbat

Distributed data-parallel algorithms aim to accelerate the training of deep neural networks by parallelizing the computation of large mini-batch gradient updates across multiple nodes.

General Classification Image Classification +2

On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length

1 code implementation ICLR 2019 Stanisław Jastrzębski, Zachary Kenton, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey

When studying the SGD dynamics in relation to the sharpest directions in this initial phase, we find that the SGD step is large compared to the curvature and commonly fails to minimize the loss along the sharpest directions.

Fast Approximate Natural Gradient Descent in a Kronecker-factored Eigenbasis

4 code implementations11 Jun 2018 Thomas George, César Laurent, Xavier Bouthillier, Nicolas Ballas, Pascal Vincent

Optimization algorithms that leverage gradient covariance information, such as variants of natural gradient descent (Amari, 1998), offer the prospect of yielding more effective descent directions.

Three Factors Influencing Minima in SGD

no code implementations ICLR 2018 Stanisław Jastrzębski, Zachary Kenton, Devansh Arpit, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey

In particular we find that the ratio of learning rate to batch size is a key determinant of SGD dynamics and of the width of the final minima, and that higher values of the ratio lead to wider minima and often better generalization.

Memorization Open-Ended Question Answering

Residual Connections Encourage Iterative Inference

no code implementations ICLR 2018 Stanisław Jastrzębski, Devansh Arpit, Nicolas Ballas, Vikas Verma, Tong Che, Yoshua Bengio

In general, a Resnet block tends to concentrate representation learning behavior in the first few layers while higher layers perform iterative refinement of features.

Representation Learning

A dataset and exploration of models for understanding video data through fill-in-the-blank question-answering

2 code implementations CVPR 2017 Tegan Maharaj, Nicolas Ballas, Anna Rohrbach, Aaron Courville, Christopher Pal

In addition to presenting statistics and a description of the dataset, we perform a detailed analysis of 5 different models' predictions, and compare these with human performance.

Descriptive Language Modelling +3

Theano: A Python framework for fast computation of mathematical expressions

1 code implementation9 May 2016 The Theano Development Team, Rami Al-Rfou, Guillaume Alain, Amjad Almahairi, Christof Angermueller, Dzmitry Bahdanau, Nicolas Ballas, Frédéric Bastien, Justin Bayer, Anatoly Belikov, Alexander Belopolsky, Yoshua Bengio, Arnaud Bergeron, James Bergstra, Valentin Bisson, Josh Bleecher Snyder, Nicolas Bouchard, Nicolas Boulanger-Lewandowski, Xavier Bouthillier, Alexandre de Brébisson, Olivier Breuleux, Pierre-Luc Carrier, Kyunghyun Cho, Jan Chorowski, Paul Christiano, Tim Cooijmans, Marc-Alexandre Côté, Myriam Côté, Aaron Courville, Yann N. Dauphin, Olivier Delalleau, Julien Demouth, Guillaume Desjardins, Sander Dieleman, Laurent Dinh, Mélanie Ducoffe, Vincent Dumoulin, Samira Ebrahimi Kahou, Dumitru Erhan, Ziye Fan, Orhan Firat, Mathieu Germain, Xavier Glorot, Ian Goodfellow, Matt Graham, Caglar Gulcehre, Philippe Hamel, Iban Harlouchet, Jean-Philippe Heng, Balázs Hidasi, Sina Honari, Arjun Jain, Sébastien Jean, Kai Jia, Mikhail Korobov, Vivek Kulkarni, Alex Lamb, Pascal Lamblin, Eric Larsen, César Laurent, Sean Lee, Simon Lefrancois, Simon Lemieux, Nicholas Léonard, Zhouhan Lin, Jesse A. Livezey, Cory Lorenz, Jeremiah Lowin, Qianli Ma, Pierre-Antoine Manzagol, Olivier Mastropietro, Robert T. McGibbon, Roland Memisevic, Bart van Merriënboer, Vincent Michalski, Mehdi Mirza, Alberto Orlandi, Christopher Pal, Razvan Pascanu, Mohammad Pezeshki, Colin Raffel, Daniel Renshaw, Matthew Rocklin, Adriana Romero, Markus Roth, Peter Sadowski, John Salvatier, François Savard, Jan Schlüter, John Schulman, Gabriel Schwartz, Iulian Vlad Serban, Dmitriy Serdyuk, Samira Shabanian, Étienne Simon, Sigurd Spieckermann, S. Ramana Subramanyam, Jakub Sygnowski, Jérémie Tanguay, Gijs van Tulder, Joseph Turian, Sebastian Urban, Pascal Vincent, Francesco Visin, Harm de Vries, David Warde-Farley, Dustin J. Webb, Matthew Willson, Kelvin Xu, Lijun Xue, Li Yao, Saizheng Zhang, Ying Zhang

Since its introduction, it has been one of the most used CPU and GPU mathematical compilers - especially in the machine learning community - and has shown steady performance improvements.

BIG-bench Machine Learning Clustering +2

Recurrent Batch Normalization

3 code implementations30 Mar 2016 Tim Cooijmans, Nicolas Ballas, César Laurent, Çağlar Gülçehre, Aaron Courville

We propose a reparameterization of LSTM that brings the benefits of batch normalization to recurrent neural networks.

General Classification Language Modelling +3

Dynamic Capacity Networks

2 code implementations24 Nov 2015 Amjad Almahairi, Nicolas Ballas, Tim Cooijmans, Yin Zheng, Hugo Larochelle, Aaron Courville

The low-capacity sub-networks are applied across most of the input, but also provide a guide to select a few portions of the input on which to apply the high-capacity sub-networks.

Delving Deeper into Convolutional Networks for Learning Video Representations

2 code implementations19 Nov 2015 Nicolas Ballas, Li Yao, Chris Pal, Aaron Courville

We propose an approach to learn spatio-temporal features in videos from intermediate visual representations we call "percepts" using Gated-Recurrent-Unit Recurrent Networks (GRUs). Our method relies on percepts that are extracted from all level of a deep convolutional network trained on the large ImageNet dataset.

Action Recognition Temporal Action Localization +1

Oracle performance for visual captioning

1 code implementation14 Nov 2015 Li Yao, Nicolas Ballas, Kyunghyun Cho, John R. Smith, Yoshua Bengio

The task of associating images and videos with a natural language description has attracted a great amount of attention recently.

Image Captioning Language Modelling +1

Describing Videos by Exploiting Temporal Structure

5 code implementations ICCV 2015 Li Yao, Atousa Torabi, Kyunghyun Cho, Nicolas Ballas, Christopher Pal, Hugo Larochelle, Aaron Courville

In this context, we propose an approach that successfully takes into account both the local and global temporal structure of videos to produce descriptions.

Action Recognition Temporal Action Localization +1

FitNets: Hints for Thin Deep Nets

3 code implementations19 Dec 2014 Adriana Romero, Nicolas Ballas, Samira Ebrahimi Kahou, Antoine Chassang, Carlo Gatta, Yoshua Bengio

In this paper, we extend this idea to allow the training of a student that is deeper and thinner than the teacher, using not only the outputs but also the intermediate representations learned by the teacher as hints to improve the training process and final performance of the student.

Knowledge Distillation

Cannot find the paper you are looking for? You can Submit a new open access paper.