Publications

The Bottleneck Simulator: A Model-based Deep Reinforcement Learning Approach
Iulian Vlad Serban
Chinnadhurai Sankar
Michael Pieper
Joelle Pineau
Yoshua Bengio
Deep reinforcement learning has recently shown many impressive successes. However, one major obstacle towards applying such methods to real-… (see more)world problems is their lack of data-efficiency. To this end, we propose the Bottleneck Simulator: a model-based reinforcement learning method which combines a learned, factorized transition model of the environment with rollout simulations to learn an effective policy from few examples. The learned transition model employs an abstract, discrete (bottleneck) state, which increases sample efficiency by reducing the number of model parameters and by exploiting structural properties of the environment. We provide a mathematical analysis of the Bottleneck Simulator in terms of fixed points of the learned policy, which reveals how performance is affected by four distinct sources of error: an error related to the abstract space structure, an error related to the transition model estimation variance, an error related to the transition model estimation bias, and an error related to the transition model class bias. Finally, we evaluate the Bottleneck Simulator on two natural language processing tasks: a text adventure game and a real-world, complex dialogue response selection task. On both tasks, the Bottleneck Simulator yields excellent performance beating competing approaches.
2020-10-26
Journal of Artificial Intelligence Research (published)
doi.org
arxiv.org
Association between extreme precipitation, drinking water and acute gastrointestinal illness in the Great Lakes
R. Graydon
M. Mezzacapo
J. Boehme
David L Buckeridge
S. Foldy
T. Edge
J. Brubacher
L. Chan
M. Dellinger
E. Faustman
J. Rose
T. Takaro
2020-10-25
ISEE Conference Abstracts (published)
doi.org
DoMoBOT: a bot for automated and interactive domain modelling
Rijul Saini
Gunter Mussbacher
Jin L.C. Guo
Jörg Kienzle
Domain modelling transforms domain problem descriptions written in natural language (NL) into analyzable and concise domain models (class di… (see more)agrams) during requirements analysis or the early stages of design in software development. Since the practice of domain modelling requires time in addition to modelling skills and experience, several approaches have been proposed to automate or semi-automate the construction of domain models from problem descriptions expressed in NL. Despite the existing work on domain model extraction, some significant challenges remain unaddressed: (i) the extracted domain models are not accurate enough to be used directly or with minor modifications in software development, (ii) existing approaches do not facilitate the tracing of the rationale behind the modelling decisions taken by the model extractor, and (iii) existing approaches do not provide interactive interfaces to update the extracted domain models. Therefore, in this paper, we introduce a domain modelling bot called DoMoBOT, explain its architecture, and implement it in the form of a web-based prototype tool. The bot automatically extracts a domain model from a problem description written in NL with an accuracy higher than existing approaches. Furthermore, the bot enables modellers to update a part of the extracted domain model and in response the bot re-configures the other parts of the domain model pro-actively. To improve the accuracy of extracted domain models, we combine the techniques of Natural Language Processing and Machine Learning. Finally, we evaluate the accuracy of the extracted domain models.
2020-10-25
Proceedings of the 23rd ACM/IEEE International Conference on Model Driven Engineering Languages and Systems: Companion Proceedings (published)
doi.org
Learning Domain Randomization Distributions for Training Robust Locomotion Policies
Melissa Mozian
Juan Camilo Gamboa Higuera
David Meger
Gregory Dudek
This paper considers the problem of learning behaviors in simulation without knowledge of the precise dynamical properties of the target rob… (see more)ot platform(s). In this context, our learning goal is to mutually maximize task efficacy on each environment considered and generalization across the widest possible range of environmental conditions. The physical parameters of the simulator are modified by a component of our technique that learns the Domain Randomization (DR) that is appropriate at each learning epoch to maximally challenge the current behavior policy, without being overly challenging, which can hinder learning progress. This so-called sweet spot distribution is a selection of simulated domains with the following properties: 1) The trained policy should be successful in environments sampled from the domain randomization distribution; and 2) The DR distribution made as wide as possible, to increase variability in the environments. These properties aim to ensure the trajectories encountered in the target system are close to those observed during training, as existing methods in machine learning are better suited for interpolation than extrapolation. We show how adapting the DR distribution while training context-conditioned policies results in improvements on jump-start and asymptotic performance when transferring a learned policy to the target environment1.
2020-10-23
2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (published)
doi.org
Veille sur les outils numériques en santé dans le contexte de COVID-19
Aude Motulsky
Philippe Després
Cecile Petitgand
Jean Noel Nikiema
Catherine Régis
Jean-Louis Denis
2020-10-22
(published)
doi.org
NU-GAN: High resolution neural upsampling with GAN
Rithesh Kumar
Kundan Kumar
Vicki Anand
Yoshua Bengio
Aaron Courville
In this paper, we propose NU-GAN, a new method for resampling audio from lower to higher sampling rates (upsampling). Audio upsampling is an… (see more) important problem since productionizing generative speech technology requires operating at high sampling rates. Such applications use audio at a resolution of 44.1 kHz or 48 kHz, whereas current speech synthesis methods are equipped to handle a maximum of 24 kHz resolution. NU-GAN takes a leap towards solving audio upsampling as a separate component in the text-to-speech (TTS) pipeline by leveraging techniques for audio generation using GANs. ABX preference tests indicate that our NU-GAN resampler is capable of resampling 22 kHz to 44.1 kHz audio that is distinguishable from original audio only 7.4% higher than random chance for single speaker dataset, and 10.8% higher than chance for multi-speaker dataset.
2020-10-21
arXiv (preprint)
doi.org
arxiv.org
Explicitly Modeling Syntax in Language Model improves Generalization
Yikang Shen
Shawn Tan
Alessandro Sordoni
Siva Reddy
Aaron Courville
Syntax is fundamental to our thinking about language. Although neural networks are very successful in many tasks, they do not explicitly mod… (see more)el syntactic structure. Failing to capture the structure of inputs could lead to generalization problems and over-parametrization. In the present work, we propose a new syntax-aware language model: Syntactic Ordered Memory (SOM). The model explicitly models the structure with a one-step look-ahead parser and maintains the conditional probability setting of the standard language model. Experiments show that SOM can achieve strong results in language modeling and syntactic generalization tests, while using fewer parameters then other models.
2020-10-20
arXiv.org (preprint)
dblp.uni-trier.de
Cross-Modal Information Maximization for Medical Imaging: CMIM
Tristan Sylvain
Francis Dutil
Tess Berthier
Lisa Di Jorio
Margaux Luck
R Devon Hjelm
Yoshua Bengio
2020-10-19
ArXiv (preprint)
arxiv.org
arxiv.org
Mutations associated with neuropsychiatric conditions delineate functional brain connectivity dimensions contributing to autism and schizophrenia
Clara A. Moreau
Sebastian G. W. Urchs
Pierre Orban
Catherine Schramm
Guillaume Dumas
Aurélie Labbe
Guillaume Huguet
Élise Douard
Pierre-Olivier Quirion
Amy Lin
Leila Kushan
Stephanie Grot
David Luck
Adrianna Mendrek
Stephane Potvin
Emmanuel Stip
Thomas Bourgeron
Alan C. Evans
Carrie E. Bearden
Pierre Bellec … (see 1 more)
Sébastien Jacquemont
16p11.2 and 22q11.2 Copy Number Variants (CNVs) confer high risk for Autism Spectrum Disorder (ASD), schizophrenia (SZ), and Attention-Defic… (see more)it-Hyperactivity-Disorder (ADHD), but their impact on functional connectivity (FC) remains unclear. Here we report an analysis of resting-state FC using magnetic resonance imaging data from 101 CNV carriers, 755 individuals with idiopathic ASD, SZ, or ADHD and 1,072 controls. We characterize CNV FC-signatures and use them to identify dimensions contributing to complex idiopathic conditions. CNVs have large mirror effects on FC at the global and regional level. Thalamus, somatomotor, and posterior insula regions play a critical role in dysconnectivity shared across deletions, duplications, idiopathic ASD, SZ but not ADHD. Individuals with higher similarity to deletion FC-signatures exhibit worse cognitive and behavioral symptoms. Deletion similarities identified at the connectivity level could be related to the redundant associations observed genome-wide between gene expression spatial patterns and FC-signatures. Results may explain why many CNVs affect a similar range of neuropsychiatric symptoms.
2020-10-18
Nature Communications (published)
doi.org
Parametric models for combined failure time data from an incident cohort study and a prevalent cohort study with follow-up
James H. McVittie
David B. Wolfson
David A. Stephens
Vittorio Addona
David L Buckeridge
2020-10-11
The International Journal of Biostatistics (published)
doi.org
A Fully Tensorized Recurrent Neural Network
Charles Onu
Jacob Miller
Doina Precup
2020-10-07
ArXiv (preprint)
arxiv.org
arxiv.org
HyPyP: a Hyperscanning Python Pipeline for inter-brain connectivity analysis
Anaël Ayrolles
Florence Brun
Phoebe Chen
Amir Djalovski
Yann Beauxis
Richard Delorme
Thomas Bourgeron
Suzanne Dikker
Guillaume Dumas
Abstract The bulk of social neuroscience takes a ‘stimulus-brain’ approach, typically comparing brain responses to different types of so… (see more)cial stimuli, but most of the time in the absence of direct social interaction. Over the last two decades, a growing number of researchers have adopted a ‘brain-to-brain’ approach, exploring similarities between brain patterns across participants as a novel way to gain insight into the social brain. This methodological shift has facilitated the introduction of naturalistic social stimuli into the study design (e.g. movies) and, crucially, has spurred the development of new tools to directly study social interaction, both in controlled experimental settings and in more ecologically valid environments. Specifically, ‘hyperscanning’ setups, which allow the simultaneous recording of brain activity from two or more individuals during social tasks, has gained popularity in recent years. However, currently, there is no agreed-upon approach to carry out such ‘inter-brain connectivity analysis’, resulting in a scattered landscape of analysis techniques. To accommodate a growing demand to standardize analysis approaches in this fast-growing research field, we have developed Hyperscanning Python Pipeline, a comprehensive and easy open-source software package that allows (social) neuroscientists to carry-out and to interpret inter-brain connectivity analyses.
2020-10-07
Social Cognitive and Affective Neuroscience (published)
doi.org