Publications

CtRL-Sim: Reactive and Controllable Driving Agents with Offline Reinforcement Learning
Luke Rowe
Roger Girgis
Anthony Gosselin
Bruno Carrez
Florian Golemo
Felix Heide
Liam Paull
Christopher Pal
Evaluating autonomous vehicle stacks (AVs) in simulation typically involves replaying driving logs from real-world recorded traffic. However… (see more), agents replayed from offline data do not react to the actions of the AV, and their behaviour cannot be easily controlled to simulate counterfactual scenarios. Existing approaches have attempted to address these shortcomings by proposing methods that rely on heuristics or learned generative models of real-world data but these approaches either lack realism or necessitate costly iterative sampling procedures to control the generated behaviours. In this work, we take an alternative approach and propose CtRL-Sim, a method that leverages return-conditioned offline reinforcement learning within a physics-enhanced Nocturne simulator to efficiently generate reactive and controllable traffic agents. Specifically, we process real-world driving data through the Nocturne simulator to generate a diverse offline reinforcement learning dataset, annotated with various reward terms. With this dataset, we train a return-conditioned multi-agent behaviour model that allows for fine-grained manipulation of agent behaviours by modifying the desired returns for the various reward components. This capability enables the generation of a wide range of driving behaviours beyond the scope of the initial dataset, including those representing adversarial behaviours. We demonstrate that CtRL-Sim can efficiently generate diverse and realistic safety-critical scenarios while providing fine-grained control over agent behaviours. Further, we show that fine-tuning our model on simulated safety-critical scenarios generated by our model enhances this controllability.
2024-09-04
Conference on Robot Learning (accepted)
doi.org
proceedings.mlr.press
Towards Robust Saliency Maps
Nham Le
Arie Gurfinkel
Xujie Si
Chuqin Geng
Saliency maps are one of the most popular tools to interpret the operation of a neural network: they compute input features deemed relevant … (see more)to the final prediction, which are often subsets of pixels that are easily understandable by a human being. However, it is known that relying solely on human assessment to judge a saliency map method can be misleading. In this work, we propose a new neural network verification specification called saliency-robustness, which aims to use formal methods to prove a relationship between Vanilla Gradient (VG) -- a simple yet surprisingly effective saliency map method -- and the network's prediction: given a network, if an input
2024-09-04
ACML.org/2024/Conference (published)
proceedings.mlr.press
proceedings.mlr.press
Reputation Gaming in Crowd Technical Knowledge Sharing
Iren Mazloomzadeh
Gias Uddin
Foutse Khomh
Ashkan Sami
Stack Overflow incentive system awards users with reputation scores to ensure quality. The decentralized nature of the forum may make the in… (see more)centive system prone to manipulation. This paper offers, for the first time, a comprehensive study of the reported types of reputation manipulation scenarios that might be exercised in Stack Overflow and the prevalence of such reputation gamers by a qualitative study of 1,697 posts from meta Stack Exchange sites. We found four different types of reputation fraud scenarios, such as voting rings where communities form to upvote each other repeatedly on similar posts. We developed algorithms that enable platform managers to automatically identify these suspicious reputation gaming scenarios for review. The first algorithm identifies isolated/semi-isolated communities where probable reputation frauds may occur mostly by collaborating with each other. The second algorithm looks for sudden unusual big jumps in the reputation scores of users. We evaluated the performance of our algorithms by examining the reputation history dashboard of Stack Overflow users from the Stack Overflow website. We observed that around 60-80% of users flagged as suspicious by our algorithms experienced reductions in their reputation scores by Stack Overflow.
2024-09-03
ACM Transactions on Software Engineering and Methodology (published)
doi.org
Advancing EDGE Zones to identify spatial conservation priorities of tetrapod evolutionary history
Sebastian Pipins
Jonathan E. M. Baillie
Alex Bowmer
Laura J. Pollock
Nisha Owen
Rikki Gumbs
2024-09-02
Nature Communications (published)
doi.org
Online Convex Optimization for On-Board Routing in High-Throughput Satellites
Olivier Bélanger
Jean-Luc Lupien
Olfa Ben Yahia
Stéphane Martel
Antoine Lesage-Landry
Gunes Karabulut Kurt
The rise in low Earth orbit (LEO) satellite Internet services has led to increasing demand, often exceeding available data rates and comprom… (see more)ising the quality of service. While deploying more satellites offers a short-term fix, designing higher-performance satellites with enhanced transmission capabilities provides a more sustainable solution. Achieving the necessary high capacity requires interconnecting multiple modem banks within a satellite payload. However, there is a notable gap in research on internal packet routing within extremely high-throughput satellites. To address this, we propose a real-time optimal flow allocation and priority queue scheduling method using online convex optimization-based model predictive control. We model the problem as a multi-commodity flow instance and employ an online interior-point method to solve the routing and scheduling optimization iteratively. This approach minimizes packet loss and supports real-time rerouting with low computational overhead. Our method is tested in simulation on a next-generation extremely high-throughput satellite model, demonstrating its effectiveness compared to a reference batch optimization and to traditional methods.
2024-09-01
ArXiv (preprint)
doi.org
arxiv.org
THInC: A Theory-Driven Framework for Computational Humor Detection
Victor De Marez
Thomas Winters
Ayla Rigouts Terryn
Humor is a fundamental aspect of human communication and cognition, as it plays a crucial role in social engagement. Although theories about… (see more) humor have evolved over centuries, there is still no agreement on a single, comprehensive humor theory. Likewise, computationally recognizing humor remains a significant challenge despite recent advances in large language models. Moreover, most computational approaches to detecting humor are not based on existing humor theories. This paper contributes to bridging this long-standing gap between humor theory research and computational humor detection by creating an interpretable framework for humor classification, grounded in multiple humor theories, called THInC (Theory-driven Humor Interpretation and Classification). THInC ensembles interpretable GA2M classifiers, each representing a different humor theory. We engineered a transparent flow to actively create proxy features that quantitatively reflect different aspects of theories. An implementation of this framework achieves an F1 score of 0.85. The associative interpretability of the framework enables analysis of proxy efficacy, alignment of joke features with theories, and identification of globally contributing features. This paper marks a pioneering effort in creating a humor detection framework that is informed by diverse humor theories and offers a foundation for future advancements in theory-driven humor classification. It also serves as a first step in automatically comparing humor theories in a quantitative manner.
2024-09-01
ArXiv (preprint)
doi.org
arxiv.org
Audio Editing with Non-Rigid Text Prompts
Francesco Paissan
Zhepei Wang
Mirco Ravanaelli
Paris Smaragdis
Yusuf Cem Sübakan
In this paper, we explore audio-editing with non-rigid text edits. We show that the proposed editing pipeline is able to create audio edits … (see more)that remain faithful to the input audio. We explore text prompts that perform addition, style transfer, and in-painting. We quantitatively and qualitatively show that the edits are able to obtain results which outperform Audio-LDM, a recently released text-prompted audio generation model. Qualitative inspection of the results points out that the edits given by our approach remain more faithful to the input audio in terms of keeping the original onsets and offsets of the audio events.
2024-08-31
Interspeech 2024 (published)
doi.org
arxiv.org
Clinical Care Trajectory Assessment of Children with Congenital Diaphragmatic Hernia and Neurodevelopmental Impairment
Alexandra Dimmer
Gabriel Altit
Sabrina Beauseigle
Elena Guadagno
Louise Koclas
Katryn Paquette
Ana Sant’Anna
Adam Shapiro
Dan Poenaru
Pramod Puligandla
2024-08-31
Journal of Pediatric Surgery (published)
doi.org
Data Privacy for Record Linkage and Beyond
Shurong Lin
Eric D. Kolaczyk
In a data-driven world, two prominent research problems are record linkage and data privacy, among others. Record linkage is essential for i… (see more)mproving decision-making by integrating information of the same entities from different sources. On the other hand, data privacy research seeks to balance the need to extract accurate insights from data with the imperative to protect the privacy of the entities involved. Inevitably, data privacy issues arise in the context of record linkage. This article identifies two complementary aspects at the intersection of these two fields: (1) how to ensure privacy during record linkage and (2) how to mitigate privacy risks when releasing the analysis results after record linkage. We specifically discuss privacy-preserving record linkage, differentially private regression, and related topics.
2024-08-31
Social Science Research Network (published)
doi.org
A Joint Temporal Model for Hospitalizations and ICU Admissions Due to COVID‐19 in Quebec
Mariana Carmona‐Baez
Alexandra M. Schmidt
Shirin Golchi
David L. Buckeridge
ABSTRACT Infectious respiratory diseases have been of interest in recent years for the great burden they place on health systems, for instan… (see more)ce, the severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) that caused the global COVID‐19 pandemic. As many of these diseases might require hospitalization and even intensive care unit (ICU) admission, understanding the joint dynamics of hospitalizations and ICU admissions across time and different groups of the population remains of great importance. We aim to understand the joint evolution of hospital and ICU admissions given COVID‐19 test‐positive cases in the province of Quebec, Canada. We obtain the daily counts, by age group, on the number of confirmed COVID‐19 cases, the number of hospitalizations and the number of ICU admissions due to COVID‐19, from March 2020 through October 2021 in Quebec. We propose a joint Bayesian generalized dynamic linear model for the number of hospitalizations and ICU admissions to study their temporal trends and possible associations with sex and age group. Additionally, we use transfer functions to investigate if there is a memory effect of the number of cases on hospitalizations across the different age groups. The results suggest that there is a clear distinction in the patterns of hospitalizations and ICU admissions across age groups and that the number of cases has a persistent effect on the rate of hospitalization.
2024-08-31
Stat (published)
doi.org
Do machine learning methods Make Better predictions in pharmacoepidemiology?
Ana Paula Pena-Gralle
Mireille E. Schnitzer
Sofia-Nada Boureguaa
Félix Morin
Marc-André Legault
Caroline Sirois
Alice Dragomir
Lucie Blais
2024-08-31
Annals of Epidemiology (published)
doi.org
Predicting Five-Year All-Cause Mortality in COPD Patients Using Machine Learning
Ana Paula Pena-Gralle
Amélie Forget
Sofia-Nada Boureguaa
Marc-André Legault
Lucie Blais
2024-08-31
Annals of Epidemiology (published)
doi.org