Publications

Depth Prediction for Monocular Direct Visual Odometry
Ran Cheng
Christopher Agia
David Meger
Gregory Dudek
Depth prediction from monocular images with deep CNNs is a topic of increasing interest to the community. Advances have lead to models capab… (voir plus)le of predicting disparity maps with consistent scale, which are an acceptable prior for gradient-based direct methods. With this in consideration, we exploit depth prediction as a candidate prior for the coarse initialization, tracking, and marginalization steps of the direct visual odometry system, enabling the second-order optimizer to converge faster into a precise global minimum. In addition, the given depth prior supports large baseline stereo scenarios, maintaining robust pose estimations against challenging motion states such as in-place rotation. We further refine our pose estimation with semi-online loop closure. The experiments on KITTI demonstrate that our proposed method achieves state- of-the-art performance compared to both traditional direct visual odometry and learning-based counterparts.
2020-05-12
2020 17th Conference on Computer and Robot Vision (CRV) (publié)
doi.org
Story Forest
Bang Liu
Fred X. Han
Di Niu
Linglong Kong
Kunfeng Lai
Yu Xu
Extracting events accurately from vast news corpora and organize events logically is critical for news apps and search engines, which aim to… (voir plus) organize news information collected from the Internet and present it to users in the most sensible forms. Intuitively speaking, an event is a group of news documents that report the same news incident possibly in different ways. In this article, we describe our experience of implementing a news content organization system at Tencent to discover events from vast streams of breaking news and to evolve news story structures in an online fashion. Our real-world system faces unique challenges in contrast to previous studies on topic detection and tracking (TDT) and event timeline or graph generation, in that we (1) need to accurately and quickly extract distinguishable events from massive streams of long text documents, and (2) must develop the structures of event stories in an online manner, in order to guarantee a consistent user viewing experience. In solving these challenges, we propose Story Forest, a set of online schemes that automatically clusters streaming documents into events, while connecting related events in growing trees to tell evolving stories. A core novelty of our Story Forest system is EventX, a semi-supervised scheme to extract events from massive Internet news corpora. EventX relies on a two-layered, graph-based clustering procedure to group documents into fine-grained events. We conducted extensive evaluations based on (1) 60 GB of real-world Chinese news data, (2) a large Chinese Internet news dataset that contains 11,748 news articles with truth event labels, and (3) the 20 News Groups English dataset, through detailed pilot user experience studies. The results demonstrate the superior capabilities of Story Forest to accurately identify events and organize news text into a logical structure that is appealing to human readers.
2020-05-12
ACM Transactions on Knowledge Discovery from Data (publié)
doi.org
Active Domain Randomization
Bhairav Mehta
Manfred Diaz
Florian Golemo
Christopher Pal
Liam Paull
Domain randomization is a popular technique for improving domain transfer, often used in a zero-shot setting when the target domain is unkno… (voir plus)wn or cannot easily be used for training. In this work, we empirically examine the effects of domain randomization on agent generalization. Our experiments show that domain randomization may lead to suboptimal, high-variance policies, which we attribute to the uniform sampling of environment parameters. We propose Active Domain Randomization, a novel algorithm that learns a parameter sampling strategy. Our method looks for the most informative environment variations within the given randomization ranges by leveraging the discrepancies of policy rollouts in randomized and reference environment instances. We find that training more frequently on these instances leads to better overall agent generalization. In addition, when domain randomization and policy transfer fail, Active Domain Randomization offers more insight into the deficiencies of both the chosen parameter ranges and the learned policy, allowing for more focused debugging. Our experiments across various physics-based simulated and a real-robot task show that this enhancement leads to more robust, consistent policies.
2020-05-11
Proceedings of the Conference on Robot Learning (publié)
proceedings.mlr.press
proceedings.mlr.press
Leveraging exploration in off-policy algorithms via normalizing flows
Bogdan Mazoure
Thang Doan
Audrey Durand
R Devon Hjelm
Joelle Pineau
The ability to discover approximately optimal policies in domains with sparse rewards is crucial to applying reinforcement learning (RL) in … (voir plus)many real-world scenarios. Approaches such as neural density models and continuous exploration (e.g., Go-Explore) have been proposed to maintain the high exploration rate necessary to find high performing and generalizable policies. Soft actor-critic(SAC) is another method for improving exploration that aims to combine efficient learning via off-policy updates while maximizing the policy entropy. In this work, we extend SAC to a richer class of probability distributions (e.g., multimodal) through normalizing flows (NF) and show that this significantly improves performance by accelerating the discovery of good policies while using much smaller policy representations. Our approach, which we call SAC-NF, is a simple, efficient,easy-to-implement modification and improvement to SAC on continuous control baselines such as MuJoCo and PyBullet Roboschool domains. Finally, SAC-NF does this while being significantly parameter efficient, using as few as 5.5% the parameters for an equivalent SAC model.
2020-05-11
Proceedings of the Conference on Robot Learning (publié)
doi.org
proceedings.mlr.press
Navigation Agents for the Visually Impaired: A Sidewalk Simulator and Experiments
Martin Weiss
Simon Chamorro
Roger Girgis
Margaux Luck
Samira E. Kahou
Joseph P. Cohen
Derek Nowrouzezahrai
Doina Precup
Florian Golemo
Chris Pal
Millions of blind and visually-impaired (BVI) people navigate urban environments every day, using smartphones for high-level path-planning a… (voir plus)nd white canes or guide dogs for local information. However, many BVI people still struggle to travel to new places. In our endeavor to create a navigation assistant for the BVI, we found that existing Reinforcement Learning (RL) environments were unsuitable for the task. This work introduces SEVN, a sidewalk simulation environment and a neural network-based approach to creating a navigation agent. SEVN contains panoramic images with labels for house numbers, doors, and street name signs, and formulations for several navigation tasks. We study the performance of an RL algorithm (PPO) in this setting. Our policy model fuses multi-modal observations in the form of variable resolution images, visible text, and simulated GPS data to navigate to a goal door. We hope that this dataset, simulator, and experimental results will provide a foundation for further research into the creation of agents that can assist members of the BVI community with outdoor navigation.
2020-05-11
Proceedings of the Conference on Robot Learning (publié)
doi.org
proceedings.mlr.press
Differential neural circuitry behind autism subtypes with imbalanced social-communicative and restricted repetitive behavior symptoms
Natasha Bertelsen
Isotta Landi
Richard A.I. Bethlehem
Jakob Seidlitz
Elena Maria Busuoli
Veronica Mandelli
Eleonora Satta
Stavros Trakoshis
Bonnie Auyeung
Prantik Kundu
Eva Loth
Guillaume Dumas
Sarah Baumeister
Christian Beckmann
Sven Bölte
Thomas Bourgeron
Tony Charman
Sarah Durston
Christine Ecker
Rosemary Holt … (voir 15 de plus)
Mark Johnson
Emily J. H. Jones
Luke Mason
Andreas Meyer-Lindenberg
Carolin Moessnang
Marianne Oldehinkel
Antonio Persico
Julian Tillmann
Steven C. R. Williams
Will Spooren
Declan Murphy
Jan K. Buitelaar
Simon Baron-Cohen
Meng-Chuan Lai
Michael V. Lombardo
Social-communication (SC) and restricted repetitive behaviors (RRB) are autism diagnostic symptom domains. SC and RRB severity can markedly … (voir plus)differ within and between individuals and may be underpinned by different neural circuitry and genetic mechanisms. Modeling SC-RRB balance could help identify how neural circuitry and genetic mechanisms map onto such phenotypic heterogeneity. Here we developed a phenotypic stratification model that makes highly accurate (97-99%) out-of-sample SC=RRB, SC>RRB, and RRB>SC subtype predictions. Applying this model to resting state fMRI data from the EU-AIMS LEAP dataset (n=509), we find that while the phenotypic subtypes share many commonalities in terms of intrinsic functional connectivity, they also show subtype-specific qualitative differences compared to a typically-developing group (TD). Specifically, the somatomotor network is hypoconnected with perisylvian circuitry in SC>RRB and visual association circuitry in SC=RRB. The SC=RRB subtype also showed hyperconnectivity between medial motor and anterior salience circuitry. Genes that are highly expressed within these subtype-specific networks show a differential enrichment pattern with known ASD associated genes, indicating that such circuits are affected by differing autism-associated genomic mechanisms. These results suggest that SC-RRB imbalance subtypes share some commonalities but also express subtle differences in functional neural circuitry and the genomic underpinnings behind such circuitry.
2020-05-09
bioRxiv (prépublication)
doi.org
An Empirical Study of Human Behavioral Agents in Bandits, Contextual Bandits and Reinforcement Learning.
Baihan Lin
Guillermo Cecchi
Djallel Bouneffouf
Jenna Reinen
Irina Rish
Artificial behavioral agents are often evaluated based on their consistent behaviors and performance to take sequential actions in an enviro… (voir plus)nment to maximize some notion of cumulative reward. However, human decision making in real life usually involves different strategies and behavioral trajectories that lead to the same empirical outcome. Motivated by clinical literature of a wide range of neurological and psychiatric disorders, we propose here a more general and flexible parametric framework for sequential decision making that involves a two-stream reward processing mechanism. We demonstrated that this framework is flexible and unified enough to incorporate a family of problems spanning multi-armed bandits (MAB), contextual bandits (CB) and reinforcement learning (RL), which decompose the sequential decision making process in different levels. Inspired by the known reward processing abnormalities of many mental disorders, our clinically-inspired agents demonstrated interesting behavioral trajectories and comparable performance on simulated tasks with particular reward distributions, a real-world dataset capturing human decision-making in gambling tasks, and the PacMan game across different reward stationarities in a lifelong learning setting.
2020-05-09
(publié)
www.semanticscholar.org
Unified Models of Human Behavioral Agents in Bandits, Contextual Bandits and RL
Baihan Lin
Guillermo Cecchi
Djallel Bouneffouf
Jenna Reinen
Irina Rish
2020-05-09
ArXiv (prépublication)
arxiv.org
arxiv.org
Desirable features in a decision aid for prenatal screening – what do pregnant women and their partners think? A mixed methods pilot study
Titilayo Tatiana Agbadje
S. A. Rahimi
Mélissa Côté
Andrée-Anne Tremblay
Mariama Penda Diallo
Hélène Elidor
Alex Poulin Herron
Codjo Djignefa Djade
France Légaré
Background To help pregnant women and their partners make informed value-congruent decisions about Down syndrome prenatal screening, our te… (voir plus)am developed two successive versions of a decision aid (DAv2017 and DAv2014). We aimed to assess pregnant women and their partners’ perceptions of the usefulness of the two DAs for preparing for decision making, their relative acceptability and their most desirable features. Methods This is a mixed methods pilot study. We recruited participants of study (women and their partners) when consulting for prenatal care in three clinical sites in Quebec City. To be eligible, women had to: (a) be at least 18 years old; (b) be more than 16 weeks pregnant; or having given birth in the previous year and (c) be able to speak and write in French or English. Both women and partners were invited to give their informed consent. We collected quantitative data on the usefulness of the DAs for preparing for decision making and their relative acceptability. We developed an interview grid based on the Technology Acceptance Model and Acceptability questionnaire to explore their perceptions of the most desirable features. We performed descriptive statistics and deductive analysis. Results Overall, 23 couples and 16 individual women participated in the study. The majority of participants were between 25 and 34 years old (79% of women and 59% of partners) and highly educated (66.7% of women and 54% of partners had a university-level education). DAv2017 scored higher for usefulness for preparing for decision making (86.2 ± 13 out of 100 for DAv2017 and 77.7 ± 14 for DAv2014). For most dimensions, DAv2017 was more acceptable than DAv2014 (e.g. the amount of information was found “just right” by 80% of participants for DAv2017 against 56% for DAv2014). However, participants preferred the presentation and the values clarification exercise of DAv2014. In their opinion, neither DA presented information in a completely balanced manner. They suggested adding more information about raising Down syndrome children, replacing frequencies with percentages, different values clarification methods, and a section for the partner. Conclusions A new user-centered version of the prenatal screening DA will integrate participants’ suggestions to reflect end users’ priorities.
2020-05-07
(publié)
doi.org
Leveraging cluster backbones for improving MAP inference in statistical relational models
Mohamed Hamza Ibrahim
Christopher Pal
Gilles Pesant
2020-05-06
Annals of Mathematics and Artificial Intelligence (publié)
doi.org
Option-Critic in Cooperative Multi-Agent Systems
Jhelum Chakravorty
Nadeem Ward
Julien Roy
Maxime Chevalier-Boisvert
Sumana Basu
Andrei Lupu
Doina Precup
In this paper, we investigate learning temporal abstractions in cooperative multi-agent systems, using the options framework (Sutton et al, … (voir plus)1999). First, we address the planning problem for the decentralized POMDP represented by the multi-agent system, by introducing a \emph{common information approach}. We use the notion of \emph{common beliefs} and broadcasting to solve an equivalent centralized POMDP problem. Then, we propose the Distributed Option Critic (DOC) algorithm, which uses centralized option evaluation and decentralized intra-option improvement. We theoretically analyze the asymptotic convergence of DOC and build a new multi-agent environment to demonstrate its validity. Our experiments empirically show that DOC performs competitively against baselines and scales with the number of agents.
2020-05-04
International Joint Conference on Autonomous Agents and Multiagent Systems (publié)
doi.org
arxiv.org
A Story of Two Streams: Reinforcement Learning Models from Human Behavior and Neuropsychiatry
Baihan Lin
Guillermo Cecchi
Djallel Bouneffouf
Jenna Reinen
Irina Rish
Drawing an inspiration from behavioral studies of human decision making, we propose here a more general and flexible parametric framework fo… (voir plus)r reinforcement learning that extends standard Q-learning to a two-stream model for processing positive and negative rewards, and allows to incorporate a wide range of reward-processing biases -- an important component of human decision making which can help us better understand a wide spectrum of multi-agent interactions in complex real-world socioeconomic systems, as well as various neuropsychiatric conditions associated with disruptions in normal reward processing. From the computational perspective, we observe that the proposed Split-QL model and its clinically inspired variants consistently outperform standard Q-Learning and SARSA methods, as well as recently proposed Double Q-Learning approaches, on simulated tasks with particular reward distributions, a real-world dataset capturing human decision-making in gambling tasks, and the Pac-Man game in a lifelong learning setting across different reward stationarities.
2020-05-04
International Joint Conference on Autonomous Agents and Multiagent Systems (publié)
doi.org
arxiv.org