Portrait de Guillaume Dumas

Guillaume Dumas

Membre académique associé
Professeur agrégé, Université de Montréal
Professeur adjoint, McGill University, Département de psychiatrie
Sujets de recherche
Apprentissage automatique médical
Apprentissage par renforcement
Apprentissage profond
Biologie computationnelle
Neurosciences computationnelles
Systèmes dynamiques
Théorie de l'apprentissage automatique

Biographie

Guillaume Dumas est professeur agrégé de psychiatrie computationnelle à la Faculté de médecine de l'Université de Montréal et chercheur principal du laboratoire de psychiatrie de précision et de physiologie sociale du Centre de recherche du CHU Sainte-Justine. Il est titulaire de la chaire IVADO IA en santé mentale et chercheur-boursier junior 1 du Fonds de recherche du Québec - Santé (FRQS) dans le domaine de l’ IA en santé et de la santé numérique. En 2023, il a été retenu dans le cadre du Programme des chercheurs mondiaux CIFAR-Azrieli pour le programme de recherche Cerveau, esprit et conscience. Il a également été nommé parmi les Futurs leaders canadiens de la recherche sur le cerveau par la Fondation Brain Canada.

Il a auparavant été chercheur permanent en neurosciences et en biologie computationnelle à l'Institut Pasteur (Paris, France), ainsi que chercheur postdoctoral au Center for Complex Systems and Brain Sciences à l’Université Florida Atlantic (FAU), aux États-Unis. Il est titulaire d'un diplôme d'ingénieur en ingénierie avancée et informatique (École centrale Paris), de deux masters (physique théorique, Université Paris-Saclay; sciences cognitives, ENS/EHESS/Paris 5) et d'un doctorat en neurosciences cognitives (Sorbonne Université).

Ses recherches visent à combiner l’intelligence artificielle, les neurosciences cognitives et la médecine numérique à travers un programme interdisciplinaire suivant deux axes principaux :

- L’intelligence artificielle en santé mentale, par la création de nouveaux algorithmes pour étudier le développement de l'architecture cognitive humaine et pour fournir une médecine personnalisée en neuropsychiatrie grâce à des données allant du génome à celles des téléphones intelligents;

- Les neurosciences sociales en intelligence artificielle, par la traduction de la recherche fondamentale sur le cerveau et le formalisme des systèmes dynamiques en des modèles hybrides neurocomputationnels et d’apprentissage automatique (NeuroML) et de nouvelles architectures présentant des capacités d'apprentissage social (NeuroIA Sociale et IHM).

Étudiants actuels

Maîtrise recherche - UdeM
Visiteur de recherche indépendant - CHU Sainte Justine / Université de Montréal
Maîtrise recherche - UdeM
Superviseur⋅e principal⋅e :

Publications

Arbitrary methodological decisions skew inter-brain synchronization estimates in hyperscanning-EEG studies
Marius Zimmermann
Kathrine Schultz-Nielsen
Ivana Konvalinka
Over the past decade, hyperscanning has emerged as an important methodology to study neural processes underlying human interaction using fMR… (voir plus)I, EEG, fNIRS, and MEG. However, many methodological decisions regarding preprocessing and analysis of hyperscanning data have not yet been standardized in the hyperscanning community, yet may affect inter-brain estimates. Here we systematically investigate the effects common methodological choices can have on estimates of phase-based inter-brain synchronization (IBS) measures, using real and simulated hyperscanning (dual) EEG data. Notably, we introduce a new method to compute circular correlation (CCorr) coefficients in IBS studies, which performs more reliably in comparison to the standard approach, showing that the conventional CCorr implementation leads to large fluctuations in IBS estimates due to fluctuations in circular mean directions. Furthermore, we demonstrate how short epoch durations (of 1 second or less) can lead to inflated IBS estimates in scenarios with no strong underlying interaction. Finally, we show how signal-to-noise ratios and temporal factors may confound IBS estimates, particularly when comparing e.g., resting states with conditions involving motor actions. For each of these investigated effects, we provide recommendations for future research employing hyperscanning-EEG techniques, aimed at increasing validity and replicability of inter-brain synchronization studies.
Interpersonal neural synchrony as a mechanism of therapeutic alliance? A systematic review of hyperscanning in clinical encounters
Lena Adel
Lincoln E. Moses
Elisabeth Irvine
Kyle T. Greenway
Michael Lifshitz
Therapeutic alliance is defined as the collaborative relationship between patient and therapist. Strong therapeutic alliances have been asso… (voir plus)ciated with positive clinical outcomes, but much remains unknown about alliance and its physiological basis. Interpersonal neural synchrony (INS) – the synchronization of neural signals between interacting individuals – is emerging as a novel lens for studying interactions with relatively objective and time-sensitive neuroimaging techniques. We searched four databases to identify studies of INS in clinical encounters. Our search yielded 161 articles, 46 met criteria for full-text review and 12 were included. The included articles reported INS across a total of 174 dyads, all published since 2018. Despite diverse methodologies, INS was observed in all studies and was often related to therapeutic outcomes. However, results were mixed regarding associations between INS and therapeutic alliance. These results highlight that, while promising, additional research is needed to elucidate the relationship between INS, therapeutic alliance and clinical outcomes. Future studies should aim to standardize methodologies, explore temporal dynamics (e.g., through longitudinal assessments), and include larger sample sizes.
Introducing Brain Foundation Models
Hena Ghonia
Bruno Aristimunha
Md Rifat Arefin
Sylvain Chevallier
Brain function represents one of the most complex systems driving our world. Decoding its signals poses significant challenges, particularly… (voir plus) due to the limited availability of data and the high cost of recordings. The existence of large hospital datasets and laboratory collections partially mitigates this issue. However, the lack of standardized recording protocols, varying numbers of channels, diverse setups, scenarios, and recording devices further complicate the task. This work addresses these challenges by introducing the Brain Foundation Model (BFM), a suite of open-source models trained on brain signals. These models serve as foundational tools for various types of time-series neuroimaging tasks. This work presents the first model of the BFM series, which is trained on electroencephalogram signal data. Our results demonstrate that BFM-EEG can generate signals more accurately than other models. Upon acceptance, we will release the model weights and pipeline.
LLMs and Personalities: Inconsistencies Across Scales
This study investigates the application of human psychometric assessments to large language models (LLMs) to examine their consistency and m… (voir plus)alleability in exhibiting personality traits. We administered the Big Five Inventory (BFI) and the Eysenck Personality Questionnaire-Revised (EPQ-R) to various LLMs across different model sizes and persona prompts. Our results reveal substantial variability in responses due to question order shuffling, challenging the notion of a stable LLM "personality." Larger models demonstrated more consistent responses, while persona prompts significantly influenced trait scores. Notably, the assistant persona led to more predictable scaling, with larger models exhibiting more socially desirable and less variable traits. In contrast, non-conventional personas displayed unpredictable behaviors, sometimes extending personality trait scores beyond the typical human range. These findings have important implications for understanding LLM behavior under different conditions and reflect on the consequences of scaling.
Long-term outcomes of critically ill patients with hematological malignancies: what is the impact of the coronavirus disease 2019 pandemic? Author's reply
Laveena Munshi
Sangeeta Mehta
Diagnostic tests for infections in critically ill immunocompromised patients
Adrien Joseph
Lara Zafrani
Non-invasive electroencephalography in awake cats: Feasibility and application to sensory processing in chronic pain
Aliénor Delsart
Aude Castel
Colombe Otis
Mathieu Lachance
Maude Barbeau-Grégoire
Bertrand Lussier
Franck Péron
Marc Hébert
Nicolas Lapointe
Maxim Moreau
Johanne Martel-Pelletier
Jean-Pierre Pelletier
Eric Troncy
Oxygen thresholds in critically ill patients: need for personalized targets. Author's reply.
Laveena Munshi
Challenges in multi-task learning for fMRI-based diagnosis: Benefits for psychiatric conditions and CNVs would likely require thousands of patients
Annabelle Harvey
Clara A. Moreau
Kuldeep Kumar
Sebastian G.W. Urchs
Hanad Sharmarke
Khadije Jizi
Charles-Olivier Martin
Nadine Younis
Petra Tamer
Jean-Louis Martineau
Pierre Orban
Ana Isabel Silva
Jeremy Hall
Marianne B.M. van den Bree
Michael J. Owen
David E.J. Linden
Sarah Lippé
Carrie E. Bearden
Sébastien Jacquemont
Pierre Bellec
There is a growing interest in using machine learning (ML) models to perform automatic diagnosis of psychiatric conditions; however, general… (voir plus)ising the prediction of ML models to completely independent data can lead to sharp decrease in performance. Patients with different psychiatric diagnoses have traditionally been studied independently, yet there is a growing recognition of neuroimaging signatures shared across them as well as rare genetic copy number variants (CNVs). In this work, we assess the potential of multi-task learning (MTL) to improve accuracy by characterising multiple related conditions with a single model, making use of information shared across diagnostic categories and exposing the model to a larger and more diverse dataset. As a proof of concept, we first established the efficacy of MTL in a context where there is clearly information shared across tasks: the same target (age or sex) is predicted at different sites of data collection in a large functional magnetic resonance imaging (fMRI) dataset compiled from multiple studies. MTL generally led to substantial gains relative to independent prediction at each site. Performing scaling experiments on the UK Biobank, we observed that performance was highly dependent on sample size: for large sample sizes (N > 6000) sex prediction was better using MTL across three sites (N = K per site) than prediction at a single site (N = 3K), but for small samples (N 500) MTL was actually detrimental for age prediction. We then used established machine-learning methods to benchmark the diagnostic accuracy of each of the 7 CNVs (N = 19–103) and 4 psychiatric conditions (N = 44–472) independently, replicating the accuracy previously reported in the literature on psychiatric conditions. We observed that MTL hurt performance when applied across the full set of diagnoses, and complementary analyses failed to identify pairs of conditions which would benefit from MTL. Taken together, our results show that if a successful multi-task diagnostic model of psychiatric conditions were to be developed with resting-state fMRI, it would likely require datasets with thousands of patients across different diagnoses.
Scalable Approaches for a Theory of Many Minds
A major challenge as we move towards building agents for real-world problems, which could involve a massive number of human and/or machine a… (voir plus)gents, is that we must learn to reason about the behavior of these many other agents. In this paper, we consider the problem of scaling a predictive Theory of Mind (ToM) model to a very large number of interacting agents with a fixed computational budget. Motivated by the limited diversity of agent types, existing approaches to scalable TOM learn versatile single-agent representations for quickly adapting to new agents encountered sequentially. We consider the more general setting that many agents are observed in parallel and formulate the corresponding Theory of Many Minds (ToMM) problem of estimating the joint policy. We frame the scaling behavior of solutions in terms of parameter sharing schemes and in particular propose two parameter-free architectural features that endow models with the ability to exploit action correlations: encoding a multi-agent context, and decoding through an abstracted joint action space. The increased predictive capabilities that have come with foundation models have made it easier to imagine the possibility of using these models to make simulations that imitate the behavior of many agents within complex real-world systems. Being able to perform these simulations in a general-purpose way would not only help make more capable agents, it also would be a very useful capability for applications in social science, political science, and economics.
Lost in Translation: The Algorithmic Gap Between LMs and the Brain
Tosato Tommaso
Tikeng Notsawo Pascal Junior
Helbling Saskia
Language Models (LMs) have achieved impressive performance on various linguistic tasks, but their relationship to human language processing … (voir plus)in the brain remains unclear. This paper examines the gaps and overlaps between LMs and the brain at different levels of analysis, emphasizing the importance of looking beyond input-output behavior to examine and compare the internal processes of these systems. We discuss how insights from neuroscience, such as sparsity, modularity, internal states, and interactive learning, can inform the development of more biologically plausible language models. Furthermore, we explore the role of scaling laws in bridging the gap between LMs and human cognition, highlighting the need for efficiency constraints analogous to those in biological systems. By developing LMs that more closely mimic brain function, we aim to advance both artificial intelligence and our understanding of human cognition.
Bringing together multimodal and multilevel approaches to study the emergence of social bonds between children and improve social AI
Julie Bonnaire
Justine Cassell
This protocol paper outlines an innovative multimodal and multilevel approach to studying the emergence and evolution of how children build … (voir plus)social bonds with their peers, and its potential application to improving social artificial intelligence (AI). We detail a unique hyperscanning experimental framework utilizing functional near-infrared spectroscopy (fNIRS) to observe inter-brain synchrony in child dyads during collaborative tasks and social interactions. Our proposed longitudinal study spans middle childhood, aiming to capture the dynamic development of social connections and cognitive engagement in naturalistic settings. To do so we bring together four kinds of data: the multimodal conversational behaviors that dyads of children engage in, evidence of their state of interpersonal rapport, collaborative performance on educational tasks, and inter-brain synchrony. Preliminary pilot data provide foundational support for our approach, indicating promising directions for identifying neural patterns associated with productive social interactions. The planned research will explore the neural correlates of social bond formation, informing the creation of a virtual peer learning partner in the field of Social Neuroergonomics. This protocol promises significant contributions to understanding the neural basis of social connectivity in children, while also offering a blueprint for designing empathetic and effective social AI tools, particularly for educational contexts.