AMLab Papers at ICML 2022
AMLab members will be presenting 8 papers at ICML 2022 main conference:
-
Equivariant diffusion for molecule generation in 3d (oral)
Hoogeboom, Emiel, Satorras, Vı́ctor Garcia, Vignac, Clément, and Welling, Max
[Conference page, PDF, GitHub] -
Lie Point Symmetry Data Augmentation for Neural PDE Solvers (spotlight)
Brandstetter, Johannes, Welling, Max, and Worrall, Daniel E
[Conference page, PDF, GitHub] -
Exploiting Redundancy: Separable Group Convolutional Networks on Lie Groups (spotlight)
Knigge, David M, Romero, David W, and Bekkers, Erik J
[Conferende page, PDF, GitHub] -
CITRIS: Causal Identifiability from Temporal Intervened Sequences (spotlight)
Lippe, Phillip, Magliacane, Sara, Löwe, Sindy, Asano, Yuki M, Cohen, Taco, and Gavves, Efstratios
[Conferende page, PDF, GitHub] -
Learning Symmetric Embeddings for Equivariant World Models (spotlight)
Park, Jung Yeon, Biza, Ondrej, Zhao, Linfeng, Meent, Jan-Willem, and Walters, Robin
[Conferende page, PDF] -
Adapting the Linearised Laplace Model Evidence for Modern Deep Learning (spotlight)
Antoran, Javier, Janz, David, Allingham, James Urquhart, Daxberger, Erik, Barbano, Riccardo, Nalisnick, Eric, and Hernandez-Lobato, Jose Miguel
[Conferende page, PDF] -
Calibrated Learning to Defer with One-vs-All Classifiers (spotlight)
Verma, Rajeev, and Nalisnick, Eric
[Conferende page, PDF] -
Model-based Meta Reinforcement Learning using Graph Structured Surrogate Models and Amortized Policy Search (spotlight)
Wang, Q., and Hoof, H.
[Conferende page, PDF]