Publications

Contributions of network structure, chemoarchitecture and diagnostic categories to transitions between cognitive topographies
Andrea I. Luppi
S. Parker Singleton
Justine Y. Hansen
Keith W. Jamison
Danilo Bzdok
Amy Kuceyeski
Richard F. Betzel
Bratislav Misic
The mechanisms linking the brain’s network structure to cognitively relevant activation patterns remain largely unknown. Here, by leveragi… (voir plus)ng principles of network control, we show how the architecture of the human connectome shapes transitions between 123 experimentally defined cognitive activation maps (cognitive topographies) from the NeuroSynth meta-analytic database. Specifically, we systematically integrated large-scale multimodal neuroimaging data from functional magnetic resonance imaging, diffusion tractography, cortical morphometry and positron emission tomography to simulate how anatomically guided transitions between cognitive states can be reshaped by neurotransmitter engagement or by changes in cortical thickness. Our model incorporates neurotransmitter-receptor density maps (18 receptors and transporters) and maps of cortical thickness pertaining to a wide range of mental health, neurodegenerative, psychiatric and neurodevelopmental diagnostic categories (17,000 patients and 22,000 controls). The results provide a comprehensive look-up table charting how brain network organization and chemoarchitecture interact to manifest different cognitive topographies, and establish a principled foundation for the systematic identification of ways to promote selective transitions between cognitive topographies.
2024-08-04
Nature Biomedical Engineering (publié)
doi.org
Critical dynamics in spontaneous EEG predict anesthetic-induced loss of consciousness and perturbational complexity
Charlotte Maschke
Jordan O'Byrne
Michele Angelo Colombo
Melanie Boly
Olivia Gosseries
Steven Laureys
Mario Rosanova
Karim Jerbi
Stefanie Blain-Moraes
Consciousness has been proposed to be supported by electrophysiological patterns poised at criticality, a dynamical regime which exhibits ad… (voir plus)aptive computational properties, maximally complex patterns and divergent sensitivity to perturbation. Here, we investigate dynamical properties of the resting-state electroencephalogram (EEG) of healthy subjects undergoing general anesthesia with propofol, xenon or ketamine. Importantly, all participants were unresponsive under anesthesia, while consciousness was retained only during ketamine anesthesia (in the form of vivid dreams), enabling an experimental dissociation between unresponsiveness and unconsciousness. For each condition, we measure (i) avalanche criticality, (ii) chaoticity, and (iii) criticality-related metrics, revealing that states of unconsciousness are characterized by a distancing from both avalanche criticality and the edge of chaos. We then ask whether these same dynamical properties are predictive of the perturbational complexity index (PCI), a TMS-based measure that has shown remarkably high sensitivity in detecting consciousness independently of behavior. We successfully predict individual subjects’ PCI values with considerably high accuracy from resting-state EEG dynamical properties alone. Our results establish a firm link between perturbational complexity and criticality, and provide further evidence that criticality is a necessary condition for the emergence of consciousness.
2024-08-04
Communications Biology (publié)
doi.org
Learning Hybrid Interpretable Models: Theory, Taxonomy, and Methods
Julien Ferry
Gabriel Laberge
Ulrich Matchi Aïvodji
A hybrid model involves the cooperation of an interpretable model and a complex black box. At inference, any input of the hybrid model is as… (voir plus)signed to either its interpretable or complex component based on a gating mechanism. The advantages of such models over classical ones are two-fold: 1) They grant users precise control over the level of transparency of the system and 2) They can potentially perform better than a standalone black box since redirecting some of the inputs to an interpretable model implicitly acts as regularization. Still, despite their high potential, hybrid models remain under-studied in the interpretability/explainability literature. In this paper, we remedy this fact by presenting a thorough investigation of such models from three perspectives: Theory, Taxonomy, and Methods. First, we explore the theory behind the generalization of hybrid models from the Probably-Approximately-Correct (PAC) perspective. A consequence of our PAC guarantee is the existence of a sweet spot for the optimal transparency of the system. When such a sweet spot is attained, a hybrid model can potentially perform better than a standalone black box. Secondly, we provide a general taxonomy for the different ways of training hybrid models: the Post-Black-Box and Pre-Black-Box paradigms. These approaches differ in the order in which the interpretable and complex components are trained. We show where the state-of-the-art hybrid models Hybrid-Rule-Set and Companion-Rule-List fall in this taxonomy. Thirdly, we implement the two paradigms in a single method: HybridCORELS, which extends the CORELS algorithm to hybrid modeling. By leveraging CORELS, HybridCORELS provides a certificate of optimality of its interpretable component and precise control over transparency. We finally show empirically that HybridCORELS is competitive with existing hybrid models, and performs just as well as a standalone black box (or even better) while being partly transparent.
2024-08-04
TMLR (accepté)
doi.org
openreview.net
The effect of gestational age on short- and long-term complications following primary esophageal atresia repair
Mathias Johansen
Samuel Wasserman
Dan Poenaru
Jean Martin Laberge
Sam J. Daniel
Thomas Engelhardt
2024-08-02
Brazilian Journal of Anesthesiology (publié)
doi.org
Are self-explanations from Large Language Models faithful?
Andreas Madsen
A. Chandar
Siva Reddy
2024-07-31
Findings of the Association for Computational Linguistics ACL 2024 (publié)
doi.org
arxiv.org
Behavioral Imitation with Artificial Neural Networks Leads to Personalized Models of Brain Dynamics During Videogame Play
Anirudha Kemtur
François Paugam
Basile Pinsard
Yann Harel
Pravish Sainath
Maximilien Le Clei
Julie Boyle
Karim Jerbi
Pierre Bellec
Videogames provide a promising framework to understand brain activity in a rich, engaging, and active environment, in contrast to mostly pas… (voir plus)sive tasks currently dominating the field, such as image viewing. Analyzing videogames neuroimaging data is however challenging, and relies on time-intensive manual annotations of game events, based on somewhat arbitrary rules. Here, we introduce an innovative approach using Artificial Neural networks (ANN) and brain encoding techniques to generate activation maps associated with videogame behaviour using functional magnetic resonance imaging (fMRI). As individual behavior is highly variable across subjects in complex environments, we hypothesized that ANNs need to account for subject-specific behavior to properly capture brain dynamics. In this study, we used data collected while subjects played Shinobi III: Return of the Ninja Master (Sega, 1993), an action-platformer videogame. Using imitation learning, we trained an ANN to play the game while closely replicating the unique gameplay style of individual participants. We found that hidden layers of our imitation learning model successfully encoded task-relevant neural representations, and predicted individual brain dynamics with higher accuracy than models trained on other subjects’ gameplay. Individual-specific models also outperformed a number of baselines to predict brain activity, such as pixel inputs, or button presses. The highest correlations between layer activations and brain signals were observed in biologically plausible brain areas, i.e. somatosensory, attention, and visual networks. Our results demonstrate that training subject-specific ANNs can successfully uncover brain correlates of complex behaviour. This new method combining imitation learning, brain imaging, and videogames opens new research avenues to study decision-making and psychomotor task solving in naturalistic and complex environments.
2024-07-31
bioRxiv (prépublication)
doi.org
Investigating Failures to Generalize for Coreference Resolution Models
Ian Porada
A.R. Olteanu
Kaheer Suleman
Adam Trischler
Jackie CK Cheung
Coreference resolution models are often evaluated on multiple datasets. Datasets vary, however, in how coreference is realized -- i.e., how … (voir plus)the theoretical concept of coreference is operationalized in the dataset -- due to factors such as the choice of corpora and annotation guidelines. We investigate the extent to which errors of current coreference resolution models are associated with existing differences in operationalization across datasets (OntoNotes, PreCo, and Winogrande). Specifically, we distinguish between and break down model performance into categories corresponding to several types of coreference, including coreferring generic mentions, compound modifiers, and copula predicates, among others. This break down helps us investigate how state-of-the-art models might vary in their ability to generalize across different coreference types. In our experiments, for example, models trained on OntoNotes perform poorly on generic mentions and copula predicates in PreCo. Our findings help calibrate expectations of current coreference resolution models; and, future work can explicitly account for those types of coreference that are empirically associated with poor generalization when developing models.
2024-07-31
Findings of the Association for Computational Linguistics ACL 2024 (publié)
doi.org
arxiv.org
Knowledge Distillation for Federated Learning: a Practical Guide
Alessio Mora
Irene Tenison
Paolo Bellavista
Irina Rish
Federated Learning (FL) enables the training of Deep Learning models without centrally collecting possibly sensitive raw data. This paves th… (voir plus)e way for stronger privacy guarantees when building predictive models. The most used algorithms for FL are parameter-averaging based schemes (e.g., Federated Averaging) that, however, have well known limits: (i) Clients must implement the same model architecture; (ii) Transmitting model weights and model updates implies high communication cost, which scales up with the number of model parameters; (iii) In presence of non-IID data distributions, parameter-averaging aggregation schemes perform poorly due to client model drifts. Federated adaptations of regular Knowledge Distillation (KD) can solve and/or mitigate the weaknesses of parameter-averaging FL algorithms while possibly introducing other trade-offs. In this article, we provide a review of KD-based algorithms tailored for specific FL issues.
2024-07-31
Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence (publié)
doi.org
arxiv.org
Neural differential equations for temperature control in buildings under demand response programs
Vincent Taboga
Clement Gehring
Mathieu Le Cam
Hanane Dagdougui
Pierre-Luc Bacon
2024-07-31
Applied Energy (publié)
doi.org
Noise covariance estimation in multi-task high-dimensional linear models
Kai Tan
Gabriel Romon
Lune P Bellec
2024-07-31
Bernoulli (publié)
doi.org
arxiv.org
Understanding Decision-Time vs. Background Planning in Model-Based Reinforcement Learning
Safa Alver
Doina Precup
In model-based reinforcement learning, an agent can leverage a learned model to improve its way of behaving in different ways. Two prevalent… (voir plus) approaches are decision-time planning and background planning. In this study, we are interested in understanding under what conditions and in which settings one of these two planning styles will perform better than the other in domains that require fast responses. After viewing them through the lens of dynamic programming, we first consider the classical instantiations of these planning styles and provide theoretical results and hypotheses on which one will perform better in the pure planning, planning&learning, and transfer learning settings. We then consider the modern instantiations of these planning styles and provide hypotheses on which one will perform better in the last two of the considered settings. Lastly, we perform several illustrative experiments to empirically validate both our theoretical results and hypotheses. Overall, our findings suggest that even though decision-time planning does not perform as well as background planning in their classical instantiations, in their modern instantiations, it can perform on par or better than background planning in both the planning&learning and transfer learning settings.
2024-07-31
EWRL/2024/Workshop (accepté)
doi.org
openreview.net
Differentially Private Linear Regression With Linked Data
Shurong Lin
Elliot Paquette
Eric D. Kolaczyk
There has been increasing demand for establishing privacy-preserving methodologies for modern statistics and machine learning. Differential … (voir plus)privacy, a mathematical notion from computer science, is a rising tool offering robust privacy guarantees. Recent work focuses primarily on developing differentially private versions of individual statistical and machine learning tasks, with nontrivial upstream pre-processing typically not incorporated. An important example is when record linkage is done prior to downstream modeling. Record linkage refers to the statistical task of linking two or more data sets of the same group of entities without a unique identifier. This probabilistic procedure brings additional uncertainty to the subsequent task. In this paper, we present two differentially private algorithms for linear regression with linked data. In particular, we propose a noisy gradient method and a sufficient statistics perturbation approach for the estimation of regression coefficients. We investigate the privacy-accuracy tradeoff by providing finite-sample error bounds for the estimators, which allows us to understand the relative contributions of linkage error, estimation error, and the cost of privacy. The variances of the estimators are also discussed. We demonstrate the performance of the proposed algorithms through simulations and an application to synthetic data.
2024-07-30
Harvard Data Science Review (publié)
doi.org
arxiv.org