EPFL is present @ ICML 2022
ICML 2022 © 2022 EPFL
This year, 17 EPFL papers have been accepted to the 39th International Conference on Machine Learning (ICML). 2022 edition of ICML will be held in Baltimore, USA from July 17th - 23th 2022.
We take this opportunity to congratulate all the authors of the accepted EPFL papers.
List of the accepted EPFL papers:
Minimum Cost Intervention Design for Causal Effect Identification
Sina Akbari (École Polytechnique Fédérale de Lausanne (EPFL)) · Jalal Etesami (EPFL) · Negar Kiyavash (École Polytechnique Fédérale de Lausanne)
Byzantine Machine Learning Made Easy By Resilient Averaging of Momentums
Sadegh Farhadkhani (EPFL) · Rachid Guerraoui (EPFL) · Nirupam Gupta (EPFL) · Rafael Pinot (EPFL - Ecocloud) · John Stephan (EPFL)
A Natural Actor-Critic Framework for Zero-Sum Markov Games
Ahmet Alacaoglu (University of Wisconsin-Madison) · Luca Viano (EPFL) · Niao He (ETH Zurich) · Volkan Cevher (EPFL)
Equivariant Diffusion for Molecule Generation in 3D
Emiel Hoogeboom (University of Amsterdam) · Víctor Garcia Satorras (University of Amsterdam) · Clément Vignac (EPFL) · Max Welling (University of Amsterdam & Qualcomm)
An Equivalence Between Data Poisoning and Byzantine Gradient Attacks
Sadegh Farhadkhani (EPFL) · Rachid Guerraoui (EPFL) · Lê-Nguyên Hoang (EPFL) · Oscar Villemaud (EPFL)
Online and Consistent Correlation Clustering
Vincent Cohen-Addad (Google) · Silvio Lattanzi (Google) · Andreas Maggiori (EPFL) · Nikos Parotsidis (Google)
Fluctuations, Bias, Variance & Ensemble of Learners: Exact Asymptotics for Convex Losses in High-Dimension
Bruno Loureiro (EPFL) · Cedric Gerbelot (ENS) · Maria Refinetti (Laboratoire de Physique de l’Ecole Normale Supérieure Paris) · Gabriele Sicuro (King's College London) · FLORENT KRZAKALA (EPFL)
Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation
Pier Giuseppe Sessa (ETH Zürich) · Maryam Kamgarpour (EPFL) · Andreas Krause (ETH Zurich)
Towards Understanding Sharpness-Aware Minimization
Maksym Andriushchenko (EPFL) · Nicolas Flammarion (EPFL)
AdaGrad Avoids Saddle Points
Kimon Antonakopoulos (EPFL) · Panayotis Mertikopoulos (CNRS and Criteo AI Lab) · Georgios Piliouras (Singapore University of Technology and Design) · Xiao Wang (Singapore University of Technology and Design)
Failure and success of the spectral bias prediction for Kernel Ridge Regression: the case of low-dimensional data
Umberto M. Tomasini (EPFL) · Antonio Sclocchi (EPFL - IPHYS - PCSL) · Matthieu Wyart ()
Memory-Based Model Editing at Scale
Eric Mitchell (Stanford) · Charles Lin (Stanford) · Antoine Bosselut (EPFL) · Christopher Manning (Stanford University) · Chelsea Finn (Stanford)
Scaling up Universal Methods for Convex Optimization
Kimon Antonakopoulos (EPFL) · Dong Quan Vu (Laboratoire d'informatique de Grenoble - LIG) · Volkan Cevher (EPFL) · Kfir Levy (-) · Panayotis Mertikopoulos (CNRS and Criteo AI Lab)
Score matching enables causal discovery of nonlinear additive noise models
Paul Rolland (Ecole Polytechnique Fédérale de Lausanne) · Volkan Cevher (EPFL) · Matthäus Kleindessner (Amazon) · Chris Russell (Amazon) · Dominik Janzing (Amazon Research Tübingen) · Bernhard Schölkopf (Amazon / MPI Intelligent Systems) · Francesco Locatello (Amazon Lablet)
An initial alignment between neural network and target is needed for gradient descent to learn
Emmanuel Abbe (EPFL) · Elisabetta Cornacchia (EPFL) · Jan Hazla (EPFL) · Christopher Marquis (EPFL)
SPECTRE : Spectral Conditioning Overcomes the Expressivity Limits of One-shot Graph Generators
Karolis Martinkus (ETH Zurich) · Andreas Loukas (EPFL) · Nathanaël Perraudin (Swiss Data Science Center, ETH Zürich) · Roger Wattenhofer (ETH Zurich)
The dynamics of representation learning in shallow, non-linear autoencoders
Maria Refinetti (Laboratoire de Physique de l’Ecole Normale Supérieure Paris) · Sebastian Goldt (International School of Advanced Studies (SISSA))