David Rolnick

Membre académique principal

drolnick@mila.quebec

Chaire en IA Canada-CIFAR

Professeur adjoint, McGill University, École d'informatique

Professeur associé, Université de Montréal, Département d'informatique et de recherche opérationnelle

Site web

Google Scholar

Biographie

David Rolnick est professeur adjoint et titulaire d’une chaire en IA Canada-CIFAR à l'École d'informatique de l'Université McGill et membre académique principal de Mila – Institut québécois d’intelligence artificielle. Ses travaux portent sur les applications de l'apprentissage automatique dans la lutte contre le changement climatique. Il est cofondateur et président de Climate Change AI et codirecteur scientifique de Sustainability in the Digital Age. David Rolnick a obtenu un doctorat en mathématiques appliquées du Massachusetts Institute of Technology (MIT). Il a été chercheur postdoctoral en sciences mathématiques à la National Science Foundation (NSF), chercheur diplômé à la NSF et boursier Fulbright. Il a figuré sur la liste des « 35 innovateurs de moins de 35 ans » de la MIT Technology Review en 2021.

Étudiants actuels

Alex Hernandez-Garcia

Postdoctorat - Université de Montréal

Superviseur⋅e principal⋅e :

Collaborateur·rice de recherche - Université Paris-Saclay

Co-superviseur⋅e :

Yoshua Bengio

alexandre.duval@mila.quebec

Site web

Ali Ramlaoui

Collaborateur·rice de recherche

ali.ramlaoui@mila.quebec

Github

Amna El-Mustafa

Collaborateur·rice de recherche

amna.elmustafa@mila.quebec

Anna Viklund

Collaborateur·rice de recherche

anna.viklund@mila.quebec

Github

Arthur Ouaknine

Postdoctorat - McGill University

arthur.ouaknine@mila.quebec

Collaborateur·rice de recherche

ayush.prasad@mila.quebec

Basile Terver

Collaborateur·rice de recherche

Co-superviseur⋅e :

Yoshua Bengio

basile.terver@mila.quebec

Benjamin Akera Binen

Maîtrise recherche - McGill University

Doctorat - McGill University

catherine.villeneuve@mila.quebec

Github

Chandni Nagda

Collaborateur·rice de recherche

chandni.nagda@mila.quebec

Christina Winkler

Collaborateur·rice de recherche

christina.winkler@mila.quebec

Stagiaire de recherche - Johannes Kepler University

christina.humer@mila.quebec

Site web

Github

Google Scholar

David Alexandre Mickisch

Collaborateur·rice de recherche

david.mickisch@mila.quebec

Github

Devin Kwok

Doctorat - McGill University

devin.kwok@mila.quebec

Postdoctorat - Université de Montréal

Superviseur⋅e principal⋅e :

Yoshua Bengio

donna.vakalis@mila.quebec

Github

Google Scholar

Eya Cherif

Stagiaire de recherche

eya.cherif@mila.quebec

Felix Andreas Nahrstedt

Stagiaire de recherche - Université de Montréal

felix-andreas.nahrstedt@mila.quebec

Site web

Fagner Cunha

Collaborateur·rice alumni

fagner.cunha@mila.quebec

Doctorat - McGill University

gabriel.tseng@mila.quebec

Site web

Github

Gaurav Iyer

Maîtrise recherche - McGill University

gaurav.iyer@mila.quebec

Google Scholar

Gül Sena Altıntaş

Stagiaire de recherche - Université de Montréal

gul-sena.altintas@mila.quebec

Collaborateur·rice de recherche - Karlsruhe Institute of Technology

ilija.trajkovic@mila.quebec

Github

Juan Sebastián Cañas

Collaborateur·rice de recherche

juan-sebastian.canas-silva@mila.quebec

Doctorat - McGill University

julia.kaltenborn@mila.quebec

Collaborateur·rice de recherche

julien.boussard@mila.quebec

Marlena Reil

Stagiaire de recherche - Osnabrueck university

marlena.reil@mila.quebec

Github

Mehrab Hamidi

Maîtrise recherche - McGill University

mehrab.hamidi@mila.quebec

Github

Google Scholar

Michael Bunsen

Collaborateur·rice de recherche - McGill University

michael.bunsen@mila.quebec

Site web

Github

Mohit Anand

Collaborateur·rice de recherche - The University of Dresden, Helmholtz Centre for Environmental Research Leipzig

mohit.anand@mila.quebec

Github

Nikolaos Ioannis Bountos

Collaborateur·rice de recherche - National Observatory of Athens

nikolaos.bountos@mila.quebec

Collaborateur·rice de recherche

paula.harder@mila.quebec

Collaborateur·rice de recherche

pratinav.seth@learner.manipal.edu

Site web

Github

Google Scholar

Qidong Yang

Collaborateur·rice de recherche

qidong.yang@mila.quebec

Raesetje Sefala

Doctorat - McGill University

raesetje.sefala@mila.quebec

Site web

Ruben Cartuyvels

Collaborateur·rice de recherche - KU Leuven

ruben.cartuyvels@mila.quebec

Sebastian Hickman

Stagiaire de recherche - Cambridge University

sebastian.hickman@mila.quebec

Theo Saulus

Collaborateur·rice de recherche

Co-superviseur⋅e :

Yoshua Bengio

theo.saulus@mila.quebec

Tiffany Vlaar

Postdoctorat - McGill University

tiffany.vlaar@mila.quebec

Google Scholar

Venkatesh Ramesh

Doctorat - Université de Montréal

venkatesh.ramesh@mila.quebec

Todosijevic Viktor Todosijevic

Collaborateur·rice de recherche - RWTH Aachen University (Rheinisch-Westfälische Technische Hochschule Aachen)

Co-superviseur⋅e :

Yoshua Bengio

viktor.todosijevic@mila.quebec

Github

Yuyan Chen

Maîtrise recherche - McGill University

yuyan.chen@mila.quebec

Site web

Github

Publications

TIML: Task-Informed Meta-Learning for Agriculture

Gabriel Tseng

Hannah Kerner

David Rolnick

Labeled datasets for agriculture are extremely spatially imbalanced. When developing algorithms for data-sparse regions, a natural approach … (voir plus)is to use transfer learning from data-rich regions. While standard transfer learning approaches typically leverage only direct inputs and outputs, geospatial imagery and agricultural data are rich in metadata that can inform transfer learning algorithms, such as the spatial coordinates of data-points or the class of task being learned. We build on previous work exploring the use of meta-learning for agricultural contexts in data-sparse regions and introduce task-informed meta-learning (TIML), an augmentation to model-agnostic meta-learning which takes advantage of task-specific metadata. We apply TIML to crop type classification and yield estimation, and find that TIML significantly improves performance compared to a range of benchmarks in both contexts, across a diversity of model architectures. While we focus on tasks from agriculture, TIML could offer benefits to any meta-learning setup with task-specific metadata, such as classification of geo-tagged images and species distribution modelling.

2022-02-04

ArXiv (prépublication)

arxiv.org

Understanding the Evolution of Linear Regions in Deep Reinforcement Learning

Setareh Cohan

Nam Hee Gordon Kim

David Rolnick

Michiel van de Panne

Policies produced by deep reinforcement learning are typically characterised by their learning curves, but they remain poorly understood in … (voir plus)many other respects. ReLU-based policies result in a partitioning of the input space into piecewise linear regions. We seek to understand how observed region counts and their densities evolve during deep reinforcement learning using empirical results that span a range of continuous control tasks and policy network dimensions. Intuitively, we may expect that during training, the region density increases in the areas that are frequently visited by the policy, thereby affording fine-grained control. We use recent theoretical and empirical results for the linear regions induced by neural networks in supervised learning settings for grounding and comparison of our results. Empirically, we find that the region density increases only moderately throughout training, as measured along fixed trajectories coming from the final policy. However, the trajectories themselves also increase in length during training, and thus the region densities decrease as seen from the perspective of the current trajectory. Our findings suggest that the complexity of deep reinforcement learning policies does not principally emerge from a significant growth in the complexity of functions observed on-and-around trajectories of the policy.

openreview.net

La recherche en IA au service du monde réel

Boussole des politiques en IA

Vie étudiante et ressources

David Rolnick

Biographie

Étudiants actuels

Publications

La recherche en IA au service du monde réel

Boussole des politiques en IA

Vie étudiante et ressources

Mots-clés populaires:

David Rolnick

Biographie

Étudiants actuels

Publications