Portrait of Lune Bellec

Lune Bellec

Affiliate Member
Associate Professor, Université de Montréal, Department of Psychology
Research Topics
Computational Neuroscience
Medical Machine Learning
NeuroAI
Neuroimaging (fMRI)

Biography

I am a professor at the department of Psychology of University of Montreal and the principal investigator of the laboratory for brain simulation and exploration (SIMEXP) at the Montreal Geriatrics Institute (CRIUGM). I recently joined Mila - Quebec Artificial Intelligence Institute as affiliated member, and I supervise students in computer science (cognitive computational neuroscience) at DIRO, University of Montreal.

My main research interest is to train artificial neural networks in order to jointly mimic individual human brain activity and behaviour. To achieve that goal, I lead an intensive effort of individual data collection in neuroimaging (fMRI, MEG), the Courtois project on neuronal modelling (CNeuroMod).

I am a Senior FRQS chercheur boursier, a member of the Quebec alliance for Unifying Neuroscience and AI (UNIQUE), the leader of Digital Health at CRIUGM, co-lead of the data management platform at CRIUGM and the associate director for neuroinformatics of the Functional Neuroimaging Unit (UNF) at CRIUGM.

Current Students

Master's Research - Université de Montréal
Co-supervisor :
PhD - Université de Montréal
PhD - Université de Montréal
Principal supervisor :

Publications

Proceedings of the OHBM Open Science Room 2024
Selma Lugtmeijer
Ju-Chi Yu
Xiangzhen Kong
Janine D. Bijsterbosch
Elizabeth DuPre
Oscar Esteban
Ibrahim Faye
Seok-Jun Hong
Chuan-Peng Hu
Shella Keilholz
Chun-Chia Kung
Hyeong Hun Lee
Daniel Margulies
Cyril Pernet
Franco Pestilli
Jean-Baptiste Poline
Pradeep R. Raamana
Francesco Santini
Won Mok Shim … (see 30 more)
Paul M. Thompson
Chao-Gan Yan
Niall W. Duncan
Nikhil Bhagwat
Peter Fox
Ana Van Gulick
David N. Kennedy
Gorana Pobric
Neda Sadeghi
Nick Souter
Sandeep Panta
Isabelle van der Velpen
Tonya White
Sina Mansour L.
Qing Wang
Povilas Karvelis
Anibal S. Heinsfeld
Yu-Fang Yang
Hong Ji Kim
Nur Shahidatul Nabila Binti Ibrahim
Stefano Moia
Wei Zhang
Jessica Haigh
Rose-Marie Kouwenhoven
Terra Hyun Lee
Hurshitha Vasudevan
Yuping Yang
Subapriya Suppiah
Yi-Ju Lee
Nils Muhlert
Human-AI Alignment of Learning Trajectories in Video Games: a continual RL benchmark proposal
Yann Harel
François Paugam
We propose a design for a continual reinforcement learning (CRL) benchmark called GHAIA, centered on human-AI alignment of learning trajecto… (see more)ries in structured video game environments. Using \textit{Super Mario Bros.} as a case study, gameplay is decomposed into short, annotated scenes organized into diverse task sequences based on gameplay patterns and difficulty. Evaluation protocols measure both plasticity and stability, with flexible revisit and pacing schedules. A key innovation is the inclusion of high-resolution human gameplay data collected under controlled conditions, enabling direct comparison of human and agent learning. In addition to adapting classical CRL metrics like forgetting and backward transfer, we introduce semantic transfer metrics capturing learning over groups of scenes sharing similar game patterns. We demonstrate the feasibility of our approach on human and agent data, and discuss key aspects of the first release for community input.
Human-AI Alignment of Learning Trajectories in Video Games: a continual RL benchmark proposal
Yann Harel
François Paugam
We propose a design for a continual reinforcement learning (CRL) benchmark called GHAIA, centered on human-AI alignment of learning trajecto… (see more)ries in structured video game environments. Using \textit{Super Mario Bros.} as a case study, gameplay is decomposed into short, annotated scenes organized into diverse task sequences based on gameplay patterns and difficulty. Evaluation protocols measure both plasticity and stability, with flexible revisit and pacing schedules. A key innovation is the inclusion of high-resolution human gameplay data collected under controlled conditions, enabling direct comparison of human and agent learning. In addition to adapting classical CRL metrics like forgetting and backward transfer, we introduce semantic transfer metrics capturing learning over groups of scenes sharing similar game patterns. We demonstrate the feasibility of our approach on human and agent data, and discuss key aspects of the first release for community input.
Alignment of auditory artificial networks with massive individual fMRI brain data leads to generalisable improvements in brain encoding and downstream tasks
Maelle Freteault
Loic Tetrel
Nicolas Farrugia
Artificial neural networks trained in the field of artificial intelligence (AI) have emerged as key tools to model brain processes, sparking… (see more) the idea of aligning network representations with brain dynamics to enhance performance on AI tasks. While this concept has gained support in the visual domain, we investigate here the feasibility of creating auditory artificial neural models directly aligned with individual brain activity. This objective raises major computational challenges, as models have to be trained directly with brain data, which is typically collected at a much smaller scale than data used to train AI models. We aimed to answer two key questions: (1) Can brain alignment of auditory models lead to improved brain encoding for novel, previously unseen stimuli? (2) Can brain alignment lead to generalisable representations of auditory signals that are useful for solving a variety of complex auditory tasks? To answer these questions, we relied on two massive datasets: a deep phenotyping dataset from the Courtois neuronal modelling project, where six subjects watched four seasons (36 hours) of the Friends TV series in functional magnetic resonance imaging and the HEAR benchmark, a large battery of downstream auditory tasks. We fine-tuned SoundNet, a small pretrained convolutional neural network with ∼2.5M parameters. Aligning SoundNet with brain data from three seasons of Friends led to substantial improvement in brain encoding in the fourth season, extending beyond auditory and visual cortices. We also observed consistent performance gains on the HEAR benchmark, particularly for tasks with limited training data, where brain-aligned models performed comparably to the best-performing models regardless of size. We finally compared individual and group models, finding that individual models often matched or outperformed group models in both brain encoding and downstream task performance, highlighting the data efficiency of fine-tuning with individual brain data. Our results demonstrate the feasibility of aligning artificial neural network representations with individual brain activity during auditory processing, and suggest that this alignment is particularly beneficial for tasks with limited training data. Future research is needed to establish whether larger models can achieve even better performance and whether the observed gains extend to other tasks, particularly in the context of few shot learning.
Alignment of auditory artificial networks with massive individual fMRI brain data leads to generalizable improvements in brain encoding and downstream tasks
Maelle Freteault
Loic Tetrel
Nicolas Farrugia
Artificial neural networks trained in the field of artificial intelligence (AI) have emerged as key tools to model brain processes, sparking… (see more) the idea of aligning network representations with brain dynamics to enhance performance on AI tasks. While this concept has gained support in the visual domain, we investigate here the feasibility of creating auditory artificial neural models directly aligned with individual brain activity. This objective raises major computational challenges, as models have to be trained directly with brain data, which is typically collected at a much smaller scale than data used to train AI models. We aimed to answer two key questions: (1) Can brain alignment of auditory models lead to improved brain encoding for novel, previously unseen stimuli? (2) Can brain alignment lead to generalisable representations of auditory signals that are useful for solving a variety of complex auditory tasks? To answer these questions, we relied on two massive datasets: a deep phenotyping dataset from the Courtois neuronal modelling project, where six subjects watched four seasons (36 hours) of the Friends TV series in functional magnetic resonance imaging and the HEAR benchmark, a large battery of downstream auditory tasks. We fine-tuned SoundNet, a small pretrained convolutional neural network with ∼2.5M parameters. Aligning SoundNet with brain data from three seasons of Friends led to substantial improvement in brain encoding in the fourth season, extending beyond auditory and visual cortices. We also observed consistent performance gains on the HEAR benchmark, particularly for tasks with limited training data, where brain-aligned models performed comparably to the best-performing models regardless of size. We finally compared individual and group models, finding that individual models often matched or outperformed group models in both brain encoding and downstream task performance, highlighting the data efficiency of fine-tuning with individual brain data. Our results demonstrate the feasibility of aligning artificial neural network representations with individual brain activity during auditory processing, and suggest that this alignment is particularly beneficial for tasks with limited training data. Future research is needed to establish whether larger models can achieve even better performance and whether the observed gains extend to other tasks, particularly in the context of few shot learning.
Longitudinal reproducibility of brain and spinal cord quantitative MRI biomarkers
Mathieu Boudreau
Agah Karakuzu
Arnaud Boré
Basile Pinsard
Kiril Zelenkovski
Eva Alonso‐Ortiz
Julie Boyle
Quantitative MRI (qMRI) promises better specificity, accuracy, repeatability, and reproducibility relative to its clinically-used qualitativ… (see more)e MRI counterpart. Longitudinal reproducibility is particularly important in qMRI. The goal is to reliably quantify tissue properties that may be assessed in longitudinal clinical studies throughout disease progression or during treatment. In this work, we present the initial data release of the quantitative MRI portion of the Courtois project on neural modelling (CNeuroMod), where the brain and cervical spinal cord of six participants were scanned at regular intervals over the course of several years. This first release includes three years of data collection and up to ten sessions per participant using quantitative MRI imaging protocols (T1, magnetization transfer (MTR, MTsat), and diffusion). In the brain, T1MP2RAGE, FA, MD, and RD all exhibited high longitudinal reproducibility (intraclass correlation coefficient— ICC ≃ 1 and within-subject coefficient of variations— wCV 1%). The spinal cord cross-sectional area (CSA) computed using T2w images and T1MTsat exhibited the best longitudinal reproducibility (ICC ≃ 1 and 0.7 respectively, and wCV 2.4% and 6.9%). Results from this work show the level of longitudinal reproducibility that can be expected from qMRI protocols in the brain and spinal cord in the absence of hardware and software upgrades, and could help in the design of future longitudinal clinical studies.
Longitudinal reproducibility of brain and spinal cord quantitative MRI biomarkers
Mathieu Boudreau
Agah Karakuzu
Arnaud Boré
Basile Pinsard
Kiril Zelenkovski
Eva Alonso‐Ortiz
Julie Boyle
Abstract Quantitative MRI (qMRI) promises better specificity, accuracy, repeatability, and reproducibility relative to its clinically-used q… (see more)ualitative MRI counterpart. Longitudinal reproducibility is particularly important in qMRI. The goal is to reliably quantify tissue properties that may be assessed in longitudinal clinical studies throughout disease progression or during treatment. In this work, we present the initial data release of the quantitative MRI portion of the Courtois project on neural modelling (CNeuroMod), where the brain and cervical spinal cord of six participants were scanned at regular intervals over the course of several years. This first release includes 3 years of data collection and up to 10 sessions per participant using quantitative MRI imaging protocols (T1, magnetization transfer (MTR, MTsat), and diffusion). In the brain, T1MP2RAGE, fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD) all exhibited high longitudinal reproducibility (intraclass correlation coefficient – ICC ≃ 1 and within-subject coefficient of variations – wCV 1%). The spinal cord cross-sectional area (CSA) computed using T2w images and T1MTsat exhibited the best longitudinal reproducibility (ICC ≃ 1 and 0.7 respectively, and wCV 2.4% and 6.9%). Results from this work show the level of longitudinal reproducibility that can be expected from qMRI protocols in the brain and spinal cord in the absence of hardware and software upgrades, and could help in the design of future longitudinal clinical studies.
Longitudinal reproducibility of brain and spinal cord quantitative MRI biomarkers
Mathieu Boudreau
Agah Karakuzu
Arnaud Boré
Basile Pinsard
Kiril Zelenkovski
Eva Alonso‐Ortiz
Julie Boyle
Abstract Quantitative MRI (qMRI) promises better specificity, accuracy, repeatability, and reproducibility relative to its clinically-used q… (see more)ualitative MRI counterpart. Longitudinal reproducibility is particularly important in qMRI. The goal is to reliably quantify tissue properties that may be assessed in longitudinal clinical studies throughout disease progression or during treatment. In this work, we present the initial data release of the quantitative MRI portion of the Courtois project on neural modelling (CNeuroMod), where the brain and cervical spinal cord of six participants were scanned at regular intervals over the course of several years. This first release includes 3 years of data collection and up to 10 sessions per participant using quantitative MRI imaging protocols (T1, magnetization transfer (MTR, MTsat), and diffusion). In the brain, T1MP2RAGE, fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD) all exhibited high longitudinal reproducibility (intraclass correlation coefficient – ICC ≃ 1 and within-subject coefficient of variations – wCV 1%). The spinal cord cross-sectional area (CSA) computed using T2w images and T1MTsat exhibited the best longitudinal reproducibility (ICC ≃ 1 and 0.7 respectively, and wCV 2.4% and 6.9%). Results from this work show the level of longitudinal reproducibility that can be expected from qMRI protocols in the brain and spinal cord in the absence of hardware and software upgrades, and could help in the design of future longitudinal clinical studies.
Training Compute-Optimal Vision Transformers for Brain Encoding
Sana Ahmadi
Fraçois Paugam
Tristan Glatard
The optimal training of a vision transformer for brain encoding depends on three factors: model size, data size, and computational resources… (see more). This study investigates these three pillars, focusing on the effects of data scaling, model scaling, and high-performance computing on brain encoding results. Using VideoGPT to extract efficient spatiotemporal features from videos and training a Ridge model to predict brain activity based on these features, we conducted benchmark experiments with varying data sizes (10k, 100k, 1M, 6M) and different model configurations of GPT-2, including hidden layer dimensions, number of layers, and number of attention heads. We also evaluated the effects of training models with 32-bit vs 16-bit floating point representations. Our results demonstrate that increasing the hidden layer dimensions significantly improves brain encoding performance, as evidenced by higher Pearson correlation coefficients across all subjects. In contrast, the number of attention heads does not have a significant effect on the encoding results. Additionally, increasing the number of layers shows some improvement in brain encoding correlations, but the trend is not as consistent as that observed with hidden layer dimensions. The data scaling results show that larger training datasets lead to improved brain encoding performance, with the highest Pearson correlation coefficients observed for the largest dataset size (6M). These findings highlight that the effects of data scaling are more significant compared to model scaling in enhancing brain encoding performance. Furthermore, we explored the impact of floating-point precision by comparing 32-bit and 16-bit representations. Training with 16-bit precision yielded the same brain encoding accuracy as 32-bit, while reducing training time by 1.17 times, demonstrating its efficiency for high-performance computing tasks.
Training Compute-Optimal Vision Transformers for Brain Encoding
Sana Ahmadi
Fraçois Paugam
Tristan Glatard
The optimal training of a vision transformer for brain encoding depends on three factors: model size, data size, and computational resources… (see more). This study investigates these three pillars, focusing on the effects of data scaling, model scaling, and high-performance computing on brain encoding results. Using VideoGPT to extract efficient spatiotemporal features from videos and training a Ridge model to predict brain activity based on these features, we conducted benchmark experiments with varying data sizes (10k, 100k, 1M, 6M) and different model configurations of GPT-2, including hidden layer dimensions, number of layers, and number of attention heads. We also evaluated the effects of training models with 32-bit vs 16-bit floating point representations. Our results demonstrate that increasing the hidden layer dimensions significantly improves brain encoding performance, as evidenced by higher Pearson correlation coefficients across all subjects. In contrast, the number of attention heads does not have a significant effect on the encoding results. Additionally, increasing the number of layers shows some improvement in brain encoding correlations, but the trend is not as consistent as that observed with hidden layer dimensions. The data scaling results show that larger training datasets lead to improved brain encoding performance, with the highest Pearson correlation coefficients observed for the largest dataset size (6M). These findings highlight that the effects of data scaling are more significant compared to model scaling in enhancing brain encoding performance. Furthermore, we explored the impact of floating-point precision by comparing 32-bit and 16-bit representations. Training with 16-bit precision yielded the same brain encoding accuracy as 32-bit, while reducing training time by 1.17 times, demonstrating its efficiency for high-performance computing tasks.
Noise covariance estimation in multi-task high-dimensional linear models
Kai Tan
Gabriel Romon
A benchmark of individual auto-regressive models in a massive fMRI dataset
Fraçois Paugam
Basile Pinsard
Pierre Bellec
Dense functional magnetic resonance imaging datasets open new avenues to create auto-regressive models of brain activity. Individual idiosyn… (see more)crasies are obscured by group models, but can be captured by purely individual models given sufficient amounts of training data. In this study, we compared several deep and shallow individual models on the temporal auto-regression of BOLD time series recorded during a natural video watching task. The best performing models were then analyzed in terms of their data requirements and scaling, subject specificity and the space-time structure of their predicted dynamics. We found the Chebnets, a type of graph convolutional neural network, to be best suited for temporal BOLD auto-regression, closely followed by linear models. Chebnets demonstrated an increase in performance with increasing amounts of data, with no complete saturation at 9 h of training data. Good generalization to other kinds of video stimuli and to resting state data marked the Chebnets’ ability to capture intrinsic brain dynamics rather than only stimulus-specific autocorrelation patterns. Significant subject specificity was found at short prediction time lags. The Chebnets were found to capture lower frequencies at longer prediction time lags, and the spatial correlations in predicted dynamics were found to match traditional functional connectivity networks. Overall, these results demonstrate that large individual fMRI datasets can be used to efficiently train purely individual auto-regressive models of brain activity, and that massive amounts of individual data are required to do so. The excellent performance of the Chebnets likely reflects their ability to combine spatial and temporal interactions on large time scales at a low complexity cost. The non-linearities of the models did not appear as a key advantage. In fact, surprisingly, linear versions of the Chebnets appeared to outperform the original nonlinear ones. Individual temporal auto-regressive models have the potential to improve the predictability of the BOLD signal. This study is based on a massive, publicly-available dataset, which can serve for future benchmarks of individual auto-regressive modeling.