EPFL Papers at ICML 2021

© 2021 EPFL

© 2021 EPFL

17 EPFL papers have been accepted to the 38th International Conference on Machine Learning which will take place virtually during Jul 18th - 24th 2021. 
We take this opportunity to congratulate all the authors of the accepted EPFL papers.
If you are taking part in ICML 2021, we hope to meet and discuss!

List of accepted EPFL papers:

The Limits of Min-Max Optimization Algorithms: Convergence to Spurious Non-Critical Sets
Ya-Ping Hsieh (ETH) · Panayotis Mertikopoulos (CNRS and Criteo AI Lab) · Volkan Cevher (EPFL)

PACOH: Bayes-Optimal Meta-Learning with PAC-Guarantees
Jonas Rothfuss (ETH Zurich) · Vincent Fortuin (ETH Zürich) · Martin Josifoski (EPFL) · Andreas Krause (ETH Zurich)

Distributionally Robust Optimization with Markovian Data
Mengmeng Li (EPFL) · Tobias Sutter (EPFL) · Daniel Kuhn (EPFL)

Consensus Control for Decentralized Deep Learning
Lingjing Kong (EPFL) · Tao Lin (EPFL) · Anastasiia Koloskova (EPFL) · Martin Jaggi (EPFL) · Sebastian Stich (EPFL)

Attention is not all you need: pure attention loses rank doubly exponentially with depth
Yihe Dong (Google) · Jean-Baptiste Cordonnier (EPFL) · Andreas Loukas (EPFL)

Robust Learning-Augmented Caching: An Experimental Study
Jakub Chłędowski (Jagiellonian University) · Adam Polak (EPFL) · Bartosz Szabucki (Jagiellonian University) · Konrad Zolna (DeepMind, Jagiellonian University)

Contrastive Learning Inverts the Data Generating Process
Roland Zimmermann (University of Tübingen, International Max Planck Research School for Intelligent Systems) · Yash Sharma (University of Tübingen) · Steffen Schneider (EPFL) · Wieland Brendel (University of Tübingen) · Matthias Bethge (University of Tübingen)

Regret Minimization in Stochastic Non-Convex Learning via a Proximal-Gradient Approach
Nadav Hallak (The Technion) · Panayotis Mertikopoulos (CNRS and Criteo AI Lab) · Volkan Cevher (EPFL)

Exact Optimization of Conformal Predictors via Incremental and Decremental Learning
Giovanni Cherubin (Alan Turing Institute) · Konstantinos Chatzikokolakis (Ecole Polytechnique of Paris) · Martin Jaggi (EPFL)

Query Complexity of Adversarial Attacks
Grzegorz Gluch (EPFL) · Rüdiger Urbanke (EPFL)

Classifying high-dimensional Gaussian mixtures: Where kernel methods fail and neural networks succeed
Maria Refinetti (cole Normale Supérieure Paris) · Sebastian Goldt (IPhT) · Florent Krzakala (EPFL) · Lenka Zdeborova (EPFL)

Sequential Domain Adaptation by Synthesizing Distributionally Robust Experts
Bahar Taskesen (EPFL) · Man Chung Yue (Hong Kong Polytechnic University) · Jose Blanchet (Stanford University) · Daniel Kuhn (EPFL) · Viet Anh Nguyen (VinAI)

Cumulants of Hawkes Processes are Robust to Observation Noise
William Trouleau (EPFL) · Jalal Etesami (Bosch Research Center for AI, Germany) · Matthias Grossglauser (EPFL) · Negar Kiyavash (EPFL) · Patrick Thiran (EPFL)

Quasi-global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data
Tao Lin (EPFL) · Sai Praneeth Reddy Karimireddy (EPFL) · Sebastian Stich (EPFL) · Martin Jaggi (EPFL)

Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Berfin Simsek (EPFL) · François Ged (EPFL) · Francesco Spadaro (EPFL) · Arthur Jacot (EPFL) · Clement Hongler (EPFL) · Johanni Brea (EPFL) · Wulfram Gerstner (EPFL)

Interpretable Stability Bounds for Spectral Graph Filters
Henry Kenlay (University of Oxford) · Dorina Thanou (Swiss Data Science Center (EPFL and ETH Zurich)) · Xiaowen Dong (University of Oxford)

Learning from History for Byzantine Robust Optimization
Sai Praneeth Reddy Karimireddy (EPFL) · Lie He (EPFL) · Martin Jaggi (EPFL)