Publications

RetroGNN: Fast Estimation of Synthesizability for Virtual Screening and De Novo Design by Learning from Slow Retrosynthesis Software

Cheng-Hao Liu

Maksym Korablyov

Stanisław Jastrzębski

Paweł Włodarczyk-Pruszyński

Yoshua Bengio

Marwin Segler

2022-04-21

Journal of Chemical Information and Modeling (publié)

doi.org

Learning how to Interact with a Complex Interface using Hierarchical Reinforcement Learning

Gheorghe Comanici

Amelia Glaese

Anita Gergely

Daniel Toyama

Zafarali Ahmed

Tyler Jackson

Philippe Hamel

Doina Precup

Hierarchical Reinforcement Learning (HRL) allows interactive agents to decompose complex problems into a hierarchy of sub-tasks. Higher-leve… (voir plus)l tasks can invoke the solutions of lower-level tasks as if they were primitive actions. In this work, we study the utility of hierarchical decompositions for learning an appropriate way to interact with a complex interface. Specifically, we train HRL agents that can interface with applications in a simulated Android device. We introduce a Hierarchical Distributed Deep Reinforcement Learning architecture that learns (1) subtasks corresponding to simple finger gestures, and (2) how to combine these gestures to solve several Android tasks. Our approach relies on goal conditioning and can be used more generally to convert any base RL agent into an HRL agent. We use the AndroidEnv environment to evaluate our approach. For the experiments, the HRL agent uses a distributed version of the popular DQN algorithm to train different components of the hierarchy. While the native action space is completely intractable for simple DQN agents, our architecture can be used to establish an effective way to interact with different tasks, significantly improving the performance of the same DQN agent over different levels of abstraction.

2022-04-20

ArXiv (prépublication)

doi.org

arxiv.org

Local Learning with Neuron Groups

Adeetya Patel

Michael Eickenberg

Eugene Belilovsky

2022-04-20

ICLR.cc/2022/Workshop/Cells2Societies (poster)

doi.org

openreview.net

Summarizing Societies: Agent Abstraction in Multi-Agent Reinforcement Learning

Maximilian Puelma Touzel

Matthew D Riemer

Rupali Bhati

Irina Rish

Agents cannot make sense of many-agent societies through direct consideration of small-scale, low-level agent identities, but instead must r… (voir plus)ecognize emergent collective identities. Here, we take a first step towards a framework for recognizing this structure in large groups of low-level agents so that they can be modeled as a much smaller number of high-level agents—a process that we call agent abstraction. We illustrate this process by extending bisimulation metrics for state abstraction in reinforcement learning to the setting of multi-agent reinforcement learning and analyze a straightforward, if crude, abstraction based on experienced joint actions. It addresses non-stationarity due to other learning agents by improving minimax regret by a intuitive factor. To test if this compression factor provides signal for higher-level agency, we applied it to a large dataset of human play of the popular social dilemma game Diplomacy. We find that it correlates strongly with the degree of ground-truth abstraction of low-level units into the human players.

2022-04-20

ICLR.cc/2022/Workshop/Cells2Societies (poster)

openreview.net

An interactive meta-analysis of MRI biomarkers of myelin

Matteo Mancini

Agah Karakuzu

Julien Cohen-Adad

Mara Cercignani

Thomas E Nichols

Nikola Stikov

Several MRI measures have been proposed as in vivo biomarkers of myelin, each with applications ranging from plasticity to pathology. Despit… (voir plus)e the availability of these myelin-sensitive modalities, specificity and sensitivity have been a matter of discussion. Debate about which MRI measure is the most suitable for quantifying myelin is still ongoing. In this study, we performed a systematic review of published quantitative validation studies to clarify how different these measures are when compared to the underlying histology. We analyzed the results from 43 studies applying meta-analysis tools, controlling for study sample size and using interactive visualization ( https://neurolibre.github.io/myelin-meta-analysis ). We report the overall estimates and the prediction intervals for the coefficient of determination and find that MT and relaxometry-based measures exhibit the highest correlations with myelin content. We also show which measures are, and which measures are not statistically different regarding their relationship with histology.

2022-04-19

NeuroLibre Reproducible Preprint Server (publié)

doi.org

A Strong Node Classification Baseline for Temporal Graphs

Farimah Poursafaei

Željko Žilić

Reihaneh Rabbany

2022-04-19

Proceedings of the 2022 SIAM International Conference on Data Mining (SDM) (publié)

doi.org

Exploring age-related changes in inter-brain synchrony during verbal communication

Suzanne Dikker

Emily N. Mech

Laura Gwilliams

Tessa West

Guillaume Dumas

Kara D. Federmeier

Successful communication is key to health in older age. This is true in the narrow sense of being able to gain critical information e.g.,… (voir plus) from health care providers, but also more broadly in being able to maintain social ties and pursue meaningful activities, which, in turn, are central to maintaining health and well-being. Compared to younger adults, older adults show both quantitative and qualitative changes in how information is processed and used over time to achieve comprehension. Such systematic age-related neural dissimilarities in processing dynamics and strategies raise fundamental questions about how the human brain supports cross-generational communication, especially in light of accumulating evidence linking interpersonal similarities in brain responses to communicative success. Yet despite its prevalence and tangible health-related importance, naturalistic intergenerational communication involving older adults is understudied. In this paper, we lay out why filling this research gap is critical in advancing our understanding of naturalistic communication, with implications for both science and practice.

2022-04-18

PsyArXiv (OSF Preprints) (prépublication)

doi.org

Microscopy-BIDS: An Extension to the Brain Imaging Data Structure for Microscopy Data

Marie-Hélène Bourget

Lee Kamentsky

Satrajit S. Ghosh

Giacomo Mazzamuto

Alberto Lazari

Christopher J. Markiewicz

Robert Oostenveld

Guiomar Niso

Yaroslav O. Halchenko

Ilona Lipp

Sylvain Takerkart

Paule-Joanne Toussaint

Ali R. Khan

Gustav Nilsonne

Filippo Maria Castelli

Stefan Ross Eric Franklin Anthony Rémi Christopher J. Taylor Appelhoff

Julien Cohen-Adad

The Brain Imaging Data Structure (BIDS) is a specification for organizing, sharing, and archiving neuroimaging data and metadata in a reusab… (voir plus)le way. First developed for magnetic resonance imaging (MRI) datasets, the community-led specification evolved rapidly to include other modalities such as magnetoencephalography, positron emission tomography, and quantitative MRI (qMRI). In this work, we present an extension to BIDS for microscopy imaging data, along with example datasets. Microscopy-BIDS supports common imaging methods, including 2D/3D, ex/in vivo, micro-CT, and optical and electron microscopy. Microscopy-BIDS also includes comprehensible metadata definitions for hardware, image acquisition, and sample properties. This extension will facilitate future harmonization efforts in the context of multi-modal, multi-scale imaging such as the characterization of tissue microstructure with qMRI.

2022-04-18

Frontiers in Neuroscience (publié)

doi.org

Masked Siamese Networks for Label-Efficient Learning

Mahmoud Assran

Mathilde Caron

Ishan Misra

Piotr Bojanowski

Florian Bordes

P Vincent

Armand Joulin

Michael G. Rabbat

Nicolas Ballas

We propose Masked Siamese Networks (MSN), a self-supervised learning framework for learning image representations. Our approach matches the … (voir plus)representation of an image view containing randomly masked patches to the representation of the original unmasked image. This self-supervised pre-training strategy is particularly scalable when applied to Vision Transformers since only the unmasked patches are processed by the network. As a result, MSNs improve the scalability of joint-embedding architectures, while producing representations of a high semantic level that perform competitively on low-shot image classification. For instance, on ImageNet-1K, with only 5,000 annotated images, our base MSN model achieves 72.4% top-1 accuracy, and with 1% of ImageNet-1K labels, we achieve 75.7% top-1 accuracy, setting a new state-of-the-art for self-supervised learning on this benchmark. Our code is publicly available.

2022-04-13

ArXiv (prépublication)

doi.org

arxiv.org

Microscopy analysis neural network to solve detection, enumeration and segmentation from image-level annotations

Anthony Bilodeau

Constantin V.L. Delmas

Martin Parent

Paul De Koninck

Audrey Durand

Flavie Lavoie-Cardinal

2022-04-13