Publications

Reza Babanezhad Harikandeh

Surrogate Minimization: An Optimization Algorithm for Training Large Neural Networks with Model Parallelism

Reza Asad

Issam Hadj Laradji

Nicolas Roux

Sharan Vaswani

2023-10-25

NeurIPS.cc/2023/Workshop/OPT (poster)

The value of standards for health datasets in artificial intelligence-based applications

Anmol Arora

Joseph E. Alderman

Joanne Palmer

Shaswath Ganapathi

Elinor Laws

Melissa D. McCradden

Lauren Oakden-Rayner

Stephen R. Pfohl

Marzyeh Ghassemi

Francis McKay

Darren Treanor

Negar Rostamzadeh

Bilal Mateen

Jacqui Gath

Adewole O. Adebajo

Stephanie Kuku

Rubeta Matin

Katherine Heller

Elizabeth Sapey

Neil J. Sebire … (see 4 more)

Heather Cole-Lewis

Melanie Calvert

Alastair Denniston

Xiaoxuan Liu

2023-10-25

Nature Medicine (published)

PDB-Struct: A Comprehensive Benchmark for Structure-based Protein Design

Chuanrui Wang

Bozitao Zhong

Zuobai Zhang

Narendra Chaudhary

Sanchit Misra

Jian Tang

Structure-based protein design has attracted increasing interest, with numerous methods being introduced in recent years. However, a univers… (see more)ally accepted method for evaluation has not been established, since the wet-lab validation can be overly time-consuming for the development of new algorithms, and the

2023-10-24

NeurIPS.cc/2023/Workshop/AI4D3 (poster)

Enjeux de l’adaptation à la chaleur en ville et action publique : apports de l’interdisciplinarité et de la recherche-action - Cas de la métropole toulousaine

G. Bretagne

Julia Hidalgo

Sinda Haouès-Jouve

Guillaume Dumas

Lise Debrye

Aurélie Hanna

Valéry Masson

Le contexte législatif national, comme les attentes citoyennes exprimées pour plus d’informations et d’actions relatives aux enjeux cl… (see more)imatiques, ont progressivement incité à la territorialisation des politiques climatiques et énergétiques locales, ainsi qu’à l’émergence de l’enjeu d’adaptation climatique sur les territoires. Cette dynamique de spatialisation des enjeux climatiques trouve sa déclinaison à l’échelle de la métropole toulousaine depuis plus de 10 ans, du fait d’enjeux multiples sur le territoire : géographiques, climatiques et urbains. Les travaux de recherche menés localement autour des thématiques Ville, Environnement et Climat ont répondu au contexte favorable d’interdisciplinarité et de collaboration avec les acteurs urbains, soutenues par plusieurs appels à projets de recherche nationaux et européens. Deux objectifs majeurs sont affichés : coconstruire une connaissance afin de caractériser les enjeux climatiques et énergétiques propres au territoire toulousain, et proposer un accompagnement spécifique auprès des acteurs urbains pour mieux faire comprendre et objectiver les enjeux locaux, afin d’intégrer ces derniers dans les politiques et les actions publiques locales. Le présent article propose de revenir sur la synergie permise par cette collaboration, en s’attachant d’une part à présenter le processus de travail interdisciplinaire mis en place et, d’autre part, à montrer les productions de données et d’expertises qui en ont résulté.

2023-10-23

Climatologie (published)

Gradient Masked Averaging for Federated Learning

Irene Tenison

Sai Aravind Sreeramadas

Vaikkunth Mugunthan

Edouard Oyallon

Irina Rish

Eugene Belilovsky

Federated learning (FL) is an emerging paradigm that permits a large number of clients with heterogeneous data to coordinate learning of a u… (see more)nified global model without the need to share data amongst each other. A major challenge in federated learning is the heterogeneity of data across client, which can degrade the performance of standard FL algorithms. Standard FL algorithms involve averaging of model parameters or gradient updates to approximate the global model at the server. However, we argue that in heterogeneous settings, averaging can result in information loss and lead to poor generalization due to the bias induced by dominant client gradients. We hypothesize that to generalize better across non-i.i.d datasets, the algorithms should focus on learning the invariant mechanism that is constant while ignoring spurious mechanisms that differ across clients. Inspired from recent works in Out-of-Distribution generalization, we propose a gradient masked averaging approach for FL as an alternative to the standard averaging of client updates. This aggregation technique for client updates can be adapted as a drop-in replacement in most existing federated algorithms. We perform extensive experiments on multiple FL algorithms with in-distribution, real-world, feature-skewed out-of-distribution, and quantity imbalanced datasets and show that it provides consistent improvements, particularly in the case of heterogeneous clients.

2023-10-23

Transactions on Machine Learning Research (accepted)

Validation of ANG-1 and P-SEL as biomarkers of post-COVID-19 conditions using data from the Biobanque québécoise de la COVID-19 (BQC-19)

Eric Yamga

Antoine Soulé

Alain Piché

Amin Emad

Madeleine Durand

Simon Rousseau

2023-10-23

Clinical Proteomics (published)

Causal machine learning for single-cell genomics

Alejandro Tejada-Lapuerta

Paul Bertin

Stefan Bauer

Hananeh Aliee

Yoshua Bengio

Fabian J. Theis

2023-10-22

ArXiv (preprint)

Dosimetry of [18F]TRACK, the first PET tracer for imaging of TrkB/C receptors in humans

Alexander Thiel

Alexey Kostikov

Hailey Ahn

Youstina Daoud

Jean-Paul Soucy

Stephan Blinder

Carolin Jaworski

Carmen Wängler

Björn Wängler

Freimut Juengling

Shirin A. Enger

Ralf Schirrmacher

Background Reduced expression or impaired signalling of tropomyosin receptor kinases (Trk receptors) are found in a vast spectrum of CNS dis… (see more)orders. [^18F]TRACK is the first PET radioligand for TrkB/C with proven in vivo brain penetration and on-target specific signal. Here we report dosimetry data for [^18F]TRACK in healthy humans. 6 healthy participants (age 22–61 y, 3 female) were scanned on a General Electric Discovery PET/CT 690 scanner. [^18F]TRACK was synthesized with high molar activities (A_m = 250 ± 75 GBq/µmol), and a dynamic series of 12 whole-body scans were acquired after injection of 129 to 147 MBq of the tracer. Images were reconstructed with standard corrections using the manufacturer’s OSEM algorithm. Tracer concentration time-activity curves (TACs) were obtained using CT-derived volumes-of-interest. Organ-specific doses and the total effective dose were estimated using the Committee on Medical Internal Radiation Dose equation for adults and tabulated Source tissue values (S values). Results Average organ absorbed dose was highest for liver and gall bladder with 6.1E−2 (± 1.06E−2) mGy/MBq and 4.6 (± 1.18E−2) mGy/MBq, respectively. Total detriment weighted effective dose E_DW was 1.63E−2 ± 1.68E−3 mSv/MBq. Organ-specific TACs indicated predominantly hepatic tracer elimination. Conclusion Total and organ-specific effective doses for [^18F]TRACK are low and the dosimetry profile is similar to other ^18F-labelled radio tracers currently used in clinical settings.

2023-10-22

EJNMMI Radiopharmacy and Chemistry (published)

Hazards from Increasingly Accessible Fine-Tuning of Downloadable Foundation Models

Alan Chan

Benjamin Bucknall

Herbie Bradley

David M. Krueger

2023-10-22

NeurIPS.cc/2023/Workshop/SoLaR (spotlight)

A Novel Information-Theoretic Objective to Disentangle Representations for Fair Classification

Pierre Colombo

Nathan Noiry

Guillaume Staerman

Pablo Piantanida

2023-10-20

ArXiv (preprint)

Fundamental Limits of Membership Inference Attacks on Machine Learning Models

Eric Aubinais

Elisabeth Gassiat

Pablo Piantanida

Membership inference attacks (MIA) can reveal whether a particular data point was part of the training dataset, potentially exposing sensiti… (see more)ve information about individuals. This article provides theoretical guarantees by exploring the fundamental statistical limitations associated with MIAs on machine learning models. More precisely, we first derive the statistical quantity that governs the effectiveness and success of such attacks. We then deduce that in a very general regression setting with overfitting algorithms, attacks may have a high probability of success. Finally, we investigate several situations for which we provide bounds on this quantity of interest. Our results enable us to deduce the accuracy of potential attacks based on the number of samples and other structural parameters of learning models. In certain instances, these parameters can be directly estimated from the dataset.

2023-10-19

ArXiv (preprint)