David Rolnick

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

Aditya Aditya

Collaborateur·rice de recherche

Collaborateur·rice alumni - McGill

Collaborateur·rice de recherche - Cambridge University

Postdoctorat - McGill

Michael Bunsen

Collaborateur·rice de recherche - McGill

Shahana Chatterjee

Collaborateur·rice de recherche - N/A

Yuyan Chen

Doctorat - McGill

Eya Cherif

Collaborateur·rice de recherche - Leipzig University

Othmane Echchabi

Maîtrise recherche - McGill

Collaborateur·rice de recherche

Mohamed Elabbas

Collaborateur·rice de recherche

Jannik Endres

Collaborateur·rice de recherche

Jacopo Ghirri

Visiteur de recherche indépendant - Politecnico di Milano

Paula Harder

Visiteur de recherche indépendant

Collaborateur·rice de recherche - Johannes Kepler University

Christina Isaicu Isaicu

Collaborateur·rice de recherche - University of Amsterdam

Gaurav Iyer

Maîtrise recherche - McGill

Doctorat - McGill

Devin Kwok

Doctorat - McGill

Visiteur de recherche indépendant - Université de Montréal

Pierre-Louis Lemaire

Collaborateur·rice de recherche - Polytechnique Montréal

Superviseur⋅e principal⋅e :

Alex Hernandez-Garcia

Joshi Manoj

Collaborateur·rice de recherche - University of East Anglia

David Mickisch

Collaborateur·rice de recherche

Juan Nathaniel Nathaniel

Collaborateur·rice de recherche - Columbia university

Postdoctorat - McGill

Co-superviseur⋅e :

Lena Podina

Collaborateur·rice de recherche - University of Waterloo

Venkatesh Ramesh

Collaborateur·rice alumni - UdeM

Marlena Reil

Maîtrise recherche - McGill

Carla Roesch

Collaborateur·rice de recherche - Columbia university

Rasha Saha

Maîtrise recherche - McGill

Luca Marie Schmidt

Collaborateur·rice de recherche - University of Tübingen

Visiteur de recherche indépendant

Ilija Trajković

Collaborateur·rice de recherche - Karlsruhe Institute of Technology

Doctorat - McGill

Collaborateur·rice alumni - UdeM

Collaborateur·rice de recherche

Catherine Villeneuve

Doctorat - McGill

Shan Zhao

Collaborateur·rice de recherche - Technical University of Munich

Démocratiser l'accès aux données satellitaires grâce à l'IA

Billets de blogue

diagram illustrating how the AI foundation model for Earth observation, Galileo, works

21 octobre 2025

par

Gabriel Tseng

David Rolnick

Lire l'article

Publications

Inductive Biases for Relational Tasks

Blake Aaron Richards

Current deep learning approaches have shown good in-distribution performance but struggle in out-of-distribution settings. This is especiall… (voir plus)y true in the case of tasks involving abstract relations like recognizing rules in sequences, as required in many intelligence tests. In contrast, our brains are remarkably flexible at such tasks, an attribute that is likely linked to anatomical constraints on computations. Inspired by this, recent work has explored how enforcing that relational representations remain distinct from sensory representations can help artificial systems. Building on this work, we further explore and formalize the advantages afforded by ``partitioned'' representations of relations and sensory details. We investigate inductive biases that ensure abstract relations are learned and represented distinctly from sensory data across several neural network architectures and show that they outperform existing architectures on out-of-distribution generalization for various relational tasks. These results show that partitioning relational representations from other information streams may be a simple way to augment existing network architectures' robustness when performing relational computations.

2022-03-24

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

Tackling Climate Change with Machine Learning

Priya L. Donti

Lynn H. Kaack

Kelly Kochanski

Alexandre Lacoste

Kris Sankaran

Andrew Slavin Ross

Nikola Milojevic-Dupont

Natasha Jaques

Anna Waldman-Brown

Alexandra Luccioni

Tegan Maharaj

Evan D. Sherwin

S. Karthik Mukkavilli

Konrad P. Kording

Carla Gomes

Andrew Y. Ng

Demis Hassabis

John C. Platt

Felix Creutzig … (voir 2 de plus)

Jennifer Chayes

Yoshua Bengio

Climate change is one of the greatest challenges facing humanity, and we, as machine learning experts, may wonder how we can help. Here we d… (voir plus)escribe how machine learning can be a powerful tool in reducing greenhouse gas emissions and helping society adapt to a changing climate. From smart grids to disaster management, we identify high impact problems where existing gaps can be filled by machine learning, in collaboration with other fields. Our recommendations encompass exciting research questions as well as promising business opportunities. We call on the machine learning community to join the global effort against climate change.

2022-02-06

ACM Computing Surveys (publié)

arxiv.org

TIML: Task-Informed Meta-Learning for Agriculture

Gabriel Tseng

Hannah Kerner

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.

2022-02-03

ArXiv (prépublication)

arxiv.org

Deep ReLU Networks Preserve Expected Length

Boris Hanin

Ryan Jeong

Assessing the complexity of functions computed by a neural network helps us understand how the network will learn and generalize. One natura… (voir plus)l measure of complexity is how the network distorts length - if the network takes a unit-length curve as input, what is the length of the resulting curve of outputs? It has been widely believed that this length grows exponentially in network depth. We prove that in fact this is not the case: the expected length distortion does not grow with depth, and indeed shrinks slightly, for ReLU networks with standard random initialization. We also generalize this result by proving upper bounds both for higher moments of the length distortion and for the distortion of higher-dimensional volumes. These theoretical results are corroborated by our experiments.

2022-01-27

ICLR.cc/2022/Conference (poster)

Generating physically-consistent high-resolution climate data with hard-constrained neural networks

Paula Harder

Qidong Yang

Venkatesh Ramesh

Prasanna Sattegeri

Alex Hernandez-Garcia

Campbell Watson

D. Szwarcman

The availability of reliable, high-resolution climate and weather data is important to inform long-term decisions on climate adaptation and … (voir plus)mitigation and to guide rapid responses to extreme events. Forecasting models are limited by computational costs and therefore often can only make coarse resolution predictions. Statistical downscaling can provide an efficient method of upsampling low-resolution data. In this field, deep learning has been applied successfully, often us-ing image super-resolution methods from computer vision. Despite achieving visually compelling results in some cases, such models often violate conservation laws when predicting physical variables. In order to conserve important physical quantities, we develop methods that guarantee physical constraints are satisfied by a deep downscaling model while also increasing their performance according to traditional metrics. We introduce two ways of constraining the network: A renor-malization layer added to the end of the neural network and a successive approach that scales with increasing upsampling factors. We show the applicability of our methods across different popular architectures and upsampling factors using ERA5 reanalysis data.

2021-12-31

arXiv.org (prépublication)

TIML: Task-Informed Meta-Learning for crop type mapping

Gabriel Tseng

Hannah Kerner

Labeled datasets for agriculture are extremely spatially imbalanced. When developing algorithms for data-sparse regions, a previously explor… (voir plus)ed approach 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 is rich in metadata that can inform transfer learning algorithms, such as the spatial coordinates of data-points. We build on previous work exploring use of meta-learning to crop type mapping in data-sparse regions and introduce task-informed meta-learning (TIML), an augmentation to model-agnostic meta-learning which takes advantage of this metadata. We apply TIML to the CropHarvest dataset, a global dataset of agricultural class labels paired with remote sensing data. In addition, we introduce the concept of forgetfulness when training meta-learning models on many similar tasks to mitigate memorization of training tasks. We find that TIML significantly improves average performance across the CropHarvest evaluation tasks compared to a range of benchmark models, measured using AUC ROC and F1 scores.

2021-12-31

AAAI.org/2022/Workshop/AIAFS (publié)

Understanding the Evolution of Linear Regions in Deep Reinforcement Learning

Setareh Cohan

Nam Hee Gordon Kim

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.

2021-12-31

Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (publié)

Hidden Hypergraphs, Error-Correcting Codes, and Critical Learning in Hopfield Networks

Christopher Hillar

Tenzin Chan

Rachel Taubman

In 1943, McCulloch and Pitts introduced a discrete recurrent neural network as a model for computation in brains. The work inspired breakthr… (voir plus)oughs such as the first computer design and the theory of finite automata. We focus on learning in Hopfield networks, a special case with symmetric weights and fixed-point attractor dynamics. Specifically, we explore minimum energy flow (MEF) as a scalable convex objective for determining network parameters. We catalog various properties of MEF, such as biological plausibility, and then compare to classical approaches in the theory of learning. Trained Hopfield networks can perform unsupervised clustering and define novel error-correcting coding schemes. They also efficiently find hidden structures (cliques) in graph theory. We extend this known connection from graphs to hypergraphs and discover n-node networks with robust storage of 2Ω(n1−ϵ) memories for any ϵ>0. In the case of graphs, we also determine a critical ratio of training samples at which networks generalize completely.

2021-11-10

Entropy (publié)

Techniques for Symbol Grounding with SATNet

Sever Topan

Xujie Si

Many experts argue that the future of artificial intelligence is limited by the field's ability to integrate symbolic logical reasoning into… (voir plus) deep learning architectures. The recently proposed differentiable MAXSAT solver, SATNet, was a breakthrough in its capacity to integrate with a traditional neural network and solve visual reasoning problems. For instance, it can learn the rules of Sudoku purely from image examples. Despite its success, SATNet was shown to succumb to a key challenge in neurosymbolic systems known as the Symbol Grounding Problem: the inability to map visual inputs to symbolic variables without explicit supervision ("label leakage"). In this work, we present a self-supervised pre-training pipeline that enables SATNet to overcome this limitation, thus broadening the class of problems that SATNet architectures can solve to include datasets where no intermediary labels are available at all. We demonstrate that our method allows SATNet to attain full accuracy even with a harder problem setup that prevents any label leakage. We additionally introduce a proofreading method that further improves the performance of SATNet architectures, beating the state-of-the-art on Visual Sudoku.

2021-11-08

NeurIPS.cc/2021/Conference (spotlight)

Digitizing a sustainable future

Lucia A. Reisch

Lucas Joppa

Peter Howson

Artur Gil

Panayiota Alevizou

Nina Michaelidou

Ruby Appiah-Campbell

Tilman Santarius

Susanne Köhler

Massimo Pizzol

Pia-Johanna Schweizer

Dipti Srinivasan

Lynn H. Kaack

Priya L. Donti

2021-06-17

One Earth (publié)

ClimART: A Benchmark Dataset for Emulating Atmospheric Radiative Transfer in Weather and Climate Models

Salva Rühling Cachay

Venkatesh Ramesh

Jason N. S. Cole

Howard Barker

Numerical simulations of Earth's weather and climate require substantial amounts of computation. This has led to a growing interest in repla… (voir plus)cing subroutines that explicitly compute physical processes with approximate machine learning (ML) methods that are fast at inference time. Within weather and climate models, atmospheric radiative transfer (RT) calculations are especially expensive. This has made them a popular target for neural network-based emulators. However, prior work is hard to compare due to the lack of a comprehensive dataset and standardized best practices for ML benchmarking. To fill this gap, we build a large dataset, ClimART, with more than 10 million samples from present, pre-industrial, and future climate conditions, based on the Canadian Earth System Model. ClimART poses several methodological challenges for the ML community, such as multiple out-of-distribution test sets, underlying domain physics, and a trade-off between accuracy and inference speed. We also present several novel baselines that indicate shortcomings of datasets and network architectures used in prior work.

2020-12-31

NeurIPS Datasets and Benchmarks (publié)

DC3: A learning method for optimization with hard constraints

Priya L. Donti

J Zico Kolter

Large optimization problems with hard constraints arise in many settings, yet classical solvers are often prohibitively slow, motivating the… (voir plus) use of deep networks as cheap"approximate solvers."Unfortunately, naive deep learning approaches typically cannot enforce the hard constraints of such problems, leading to infeasible solutions. In this work, we present Deep Constraint Completion and Correction (DC3), an algorithm to address this challenge. Specifically, this method enforces feasibility via a differentiable procedure, which implicitly completes partial solutions to satisfy equality constraints and unrolls gradient-based corrections to satisfy inequality constraints. We demonstrate the effectiveness of DC3 in both synthetic optimization tasks and the real-world setting of AC optimal power flow, where hard constraints encode the physics of the electrical grid. In both cases, DC3 achieves near-optimal objective values while preserving feasibility.

2020-12-31

ICLR (publié)