no code implementations • NeurIPS 2023 • Manan Tomar, Riashat Islam, Matthew E. Taylor, Sergey Levine, Philip Bachman
We propose \textit{information gating} as a way to learn parsimonious representations that identify the minimal information required for a task.
1 code implementation • NeurIPS 2021 • Max Schwarzer, Nitarshan Rajkumar, Michael Noukhovitch, Ankesh Anand, Laurent Charlin, Devon Hjelm, Philip Bachman, Aaron Courville
Data efficiency is a key challenge for deep reinforcement learning.
Ranked #3 on Atari Games 100k on Atari 100k (using extra training data)
no code implementations • ICLR Workshop SSL-RL 2021 • Max Schwarzer, Nitarshan Rajkumar, Michael Noukhovitch, Ankesh Anand, Laurent Charlin, R Devon Hjelm, Philip Bachman, Aaron Courville
Data efficiency poses a major challenge for deep reinforcement learning.
no code implementations • 1 Jan 2021 • Alessandro Sordoni, Nouha Dziri, Hannes Schulz, Geoff Gordon, Remi Tachet des Combes, Philip Bachman
In this paper, we transform each view into a set of subviews and then decompose the original MI bound into a sum of bounds involving conditional MI between the subviews.
no code implementations • 27 Jul 2020 • R. Devon Hjelm, Philip Bachman
DeepInfoMax (DIM) is a self-supervised method which leverages the internal structure of deep networks to construct such views, forming prediction tasks between local features which depend on small patches in an image and global features which depend on the whole image.
1 code implementation • ICLR 2021 • Max Schwarzer, Ankesh Anand, Rishab Goel, R. Devon Hjelm, Aaron Courville, Philip Bachman
We further improve performance by adding data augmentation to the future prediction loss, which forces the agent's representations to be consistent across multiple views of an observation.
Ranked #9 on Atari Games 100k on Atari 100k
1 code implementation • NeurIPS 2020 • Bogdan Mazoure, Remi Tachet des Combes, Thang Doan, Philip Bachman, R. Devon Hjelm
We begin with the hypothesis that a model-free agent whose representations are predictive of properties of future states (beyond expected rewards) will be more capable of solving and adapting to new RL problems.
3 code implementations • NeurIPS 2019 • Philip Bachman, R. Devon Hjelm, William Buchwalter
Following our proposed approach, we develop a model which learns image representations that significantly outperform prior methods on the tasks we consider.
Ranked #22 on Image Classification on STL-10
1 code implementation • 7 Sep 2018 • Remi Tachet, Philip Bachman, Harm van Seijen
While recent progress has spawned very powerful machine learning systems, those agents remain extremely specialized and fail to transfer the knowledge they gain to similar yet unseen tasks.
1 code implementation • 11 Jul 2018 • Philip Bachman, Riashat Islam, Alessandro Sordoni, Zafarali Ahmed
We introduce a deep generative model for functions.
3 code implementations • ICML 2018 • Amjad Almahairi, Sai Rajeswar, Alessandro Sordoni, Philip Bachman, Aaron Courville
Learning inter-domain mappings from unpaired data can improve performance in structured prediction tasks, such as image segmentation, by reducing the need for paired data.
4 code implementations • 19 Sep 2017 • Peter Henderson, Riashat Islam, Philip Bachman, Joelle Pineau, Doina Precup, David Meger
In recent years, significant progress has been made in solving challenging problems across various domains using deep reinforcement learning (RL).
1 code implementation • 2 Aug 2017 • Philip Bachman, Doina Precup
We develop an approach to training generative models based on unrolling a variational auto-encoder into a Markov chain, and shaping the chain's trajectories using a technique inspired by recent work in Approximate Bayesian computation.
no code implementations • ICML 2017 • Philip Bachman, Alessandro Sordoni, Adam Trischler
We introduce a model that learns active learning algorithms via metalearning.
4 code implementations • WS 2017 • Xingdi Yuan, Tong Wang, Caglar Gulcehre, Alessandro Sordoni, Philip Bachman, Sandeep Subramanian, Saizheng Zhang, Adam Trischler
We propose a recurrent neural model that generates natural-language questions from documents, conditioned on answers.
1 code implementation • 6 Feb 2017 • Zihang Dai, Amjad Almahairi, Philip Bachman, Eduard Hovy, Aaron Courville
In this paper, we propose to equip Generative Adversarial Networks with the ability to produce direct energy estimates for samples. Specifically, we propose a flexible adversarial training framework, and prove this framework not only ensures the generator converges to the true data distribution, but also enables the discriminator to retain the density information at the global optimal.
Ranked #17 on Conditional Image Generation on CIFAR-10 (Inception score metric)
no code implementations • 8 Dec 2016 • Philip Bachman, Alessandro Sordoni, Adam Trischler
We develop a general problem setting for training and testing the ability of agents to gather information efficiently.
no code implementations • NeurIPS 2016 • Philip Bachman
We present an architecture which lets us train deep, directed generative models with many layers of latent variables.
2 code implementations • WS 2017 • Adam Trischler, Tong Wang, Xingdi Yuan, Justin Harris, Alessandro Sordoni, Philip Bachman, Kaheer Suleman
We present NewsQA, a challenging machine comprehension dataset of over 100, 000 human-generated question-answer pairs.
no code implementations • 11 Jun 2016 • Shikhar Sharma, Jing He, Kaheer Suleman, Hannes Schulz, Philip Bachman
Natural language generation plays a critical role in spoken dialogue systems.
1 code implementation • 7 Jun 2016 • Alessandro Sordoni, Philip Bachman, Adam Trischler, Yoshua Bengio
We propose a novel neural attention architecture to tackle machine comprehension tasks, such as answering Cloze-style queries with respect to a document.
Ranked #3 on Question Answering on Children's Book Test (Accuracy-NE metric)
no code implementations • 30 Oct 2015 • Philip Bachman, David Krueger, Doina Precup
We investigate attention as the active pursuit of useful information.
1 code implementation • NeurIPS 2015 • Philip Bachman, Doina Precup
We connect a broad class of generative models through their shared reliance on sequential decision making.
no code implementations • NeurIPS 2014 • Philip Bachman, Ouais Alsharif, Doina Precup
We formalize the notion of a pseudo-ensemble, a (possibly infinite) collection of child models spawned from a parent model by perturbing it according to some noise process.
no code implementations • 24 Feb 2014 • Ouais Alsharif, Philip Bachman, Joelle Pineau
Consider a Machine Learning Service Provider (MLSP) designed to rapidly create highly accurate learners for a never-ending stream of new tasks.