Thomas Pethick

Publication Activity (10 Years)

Years Active: 2021-2024
Publications (10 Years): 16

Top Topics

Linearly Constrained

Image Magnification

Training Process

Variational Inequalities

Top Venues

Trans. Mach. Learn. Res.

Publications

Wanyun Xie, Fabian Latorre, Kimon Antonakopoulos, Thomas Pethick, Volkan Cevher
Improving SAM Requires Rethinking its Optimization Formulation. CoRR (2024)
Wanyun Xie, Thomas Pethick, Ali Ramezani-Kebrya, Volkan Cevher
Mixed Nash for Robust Federated Learning. Trans. Mach. Learn. Res. 2024 (2024)
Ali Ramezani-Kebrya, Fanghui Liu, Thomas Pethick, Grigorios Chrysos, Volkan Cevher
Federated Learning under Covariate Shifts with Generalization Guarantees. Trans. Mach. Learn. Res. 2023 (2023)
Ali Ramezani-Kebrya, Fanghui Liu, Thomas Pethick, Grigorios Chrysos, Volkan Cevher
Federated Learning under Covariate Shifts with Generalization Guarantees. CoRR (2023)
Thomas Pethick, Olivier Fercoq, Puya Latafat, Panagiotis Patrinos, Volkan Cevher
Solving stochastic weak Minty variational inequalities without increasing batch size. CoRR (2023)
Thomas Pethick, Wanyun Xie, Volkan Cevher
Stable Nonconvex-Nonconcave Training via Linear Interpolation. CoRR (2023)
Thomas Pethick, Olivier Fercoq, Puya Latafat, Panagiotis Patrinos, Volkan Cevher
Solving stochastic weak Minty variational inequalities without increasing batch size. ICLR (2023)
Thomas Pethick, Grigorios G. Chrysos, Volkan Cevher
Revisiting adversarial training for the worst-performing class. CoRR (2023)
Thomas Pethick, Puya Latafat, Panagiotis Patrinos, Olivier Fercoq, Volkan Cevher
Escaping limit cycles: Global convergence for constrained nonconvex-nonconcave minimax problems. CoRR (2023)
Thomas Pethick, Grigorios Chrysos, Volkan Cevher
Revisiting adversarial training for the worst-performing class. Trans. Mach. Learn. Res. 2023 (2023)
Thomas Pethick, Wanyun Xie, Volkan Cevher
Stable Nonconvex-Nonconcave Training via Linear Interpolation. NeurIPS (2023)
Fabian Latorre, Igor Krawczuk, Leello Tadesse Dadi, Thomas Pethick, Volkan Cevher
Finding Actual Descent Directions for Adversarial Training. ICLR (2023)
Thomas Pethick, Puya Latafat, Panos Patrinos, Olivier Fercoq, Volkan Cevher
Escaping limit cycles: Global convergence for constrained nonconvex-nonconcave minimax problems. ICLR (2022)
Chaehwan Song, Ali Ramezani-Kebrya, Thomas Pethick, Armin Eftekhari, Volkan Cevher
Subquadratic Overparameterization for Shallow Neural Networks. NeurIPS (2021)
Kimon Antonakopoulos, Thomas Pethick, Ali Kavis, Panayotis Mertikopoulos, Volkan Cevher
Sifting through the noise: Universal first-order methods for stochastic variational inequalities. NeurIPS (2021)
Chaehwan Song, Ali Ramezani-Kebrya, Thomas Pethick, Armin Eftekhari, Volkan Cevher
Subquadratic Overparameterization for Shallow Neural Networks. CoRR (2021)