Sotabase

Career

· NSF Mathematical Sciences Postdoctoral Research Fellow, University of Pennsylvania2018–2020
· Fulbright Scholar, Freie Universität Berlin2012–2013
· Co-founder and Chair, Climate Change AI
· NSF Graduate Research Fellow, Massachusetts Institute of Technology (MIT)
· PhD in Applied Mathematics, Massachusetts Institute of Technology (MIT)
· Assistant Professor, School of Computer Science, McGill University
· Core Academic Member, Mila – Quebec Artificial Intelligence Institute
· Scientific Co-director, Sustainability in the Digital Age

Publications (131)

Deep Learning is Robust to Massive Label Noise
arXiv.org · 2017
604
cited
How to Start Training: The Effect of Initialization and Architecture
Neural Information Processing Systems · 2018
275
cited
Complexity of Linear Regions in Deep Networks
International Conference on Machine Learning · 2019
255
cited
Deep ReLU Networks Have Surprisingly Few Activation Patterns
Neural Information Processing Systems · 2019
251
cited
The power of deeper networks for expressing natural functions
International Conference on Learning Representations · 2017
187
cited
Stealing Part of a Production Language Model
International Conference on Machine Learning · 2024
141
cited
Lightweight, Pre-trained Transformers for Remote Sensing Timeseries
arXiv.org · 2023
103
cited
FAENet: Frame Averaging Equivariant GNN for Materials Modeling
International Conference on Machine Learning · 2023
87
cited
Bugs in the Data: How ImageNet Misrepresents Biodiversity
AAAI Conference on Artificial Intelligence · 2022
48
cited
A Multi-Pass Approach to Large-Scale Connectomics
arXiv.org · 2016
47
cited
Cross-Classification Clustering: An Efficient Multi-Object Tracking Technique for 3-D Instance Segmentation in Connectomics
Computer Vision and Pattern Recognition · 2018
42
cited
Hard-Constrained Deep Learning for Climate Downscaling
Journal of machine learning research · 2022
42
cited
Normalization Layers Are All That Sharpness-Aware Minimization Needs
Neural Information Processing Systems · 2023
33
cited
ClimateSet: A Large-Scale Climate Model Dataset for Machine Learning
Neural Information Processing Systems · 2023
32
cited
Fourier Neural Operators for Arbitrary Resolution Climate Data Downscaling
arXiv.org · 2023
32
cited
Hidden symmetries of ReLU networks
International Conference on Machine Learning · 2023
29
cited
On Neural Architecture Inductive Biases for Relational Tasks
arXiv.org · 2022
29
cited
Morphological Error Detection in 3D Segmentations
arXiv.org · 2017
27
cited
Generating physically-consistent high-resolution climate data with hard-constrained neural networks
arXiv.org · 2022
26
cited
Harnessing artificial intelligence to fill global shortfalls in biodiversity knowledge
Nature Reviews Biodiversity · 2025
25
cited
Sotabase
David Rolnick | Researcher Profile | Sotabase | Sotabase