LIONS @ NeurIPS 2021

7 papers from the LIONS lab have been accepted to the 2021 edition of Neural Information Processing Systems Conference (NeurIPS).

They are as follows:

1. "Sifting through the noise: Universal first-order methods for stochastic variational inequalities" by Kimon Antonakopoulos, Thomas Pethick, Ali Kavis, Panayotis Mertikopoulos and Volkan Cevher
2. "A first order primal-dual method with adaptivity to local Lipschitz smoothness" by Maria-Luiza Vladarean, Yura Malitsky, Volkan Cevher
3. "STORM+: fully adaptive SGD with Recursive Momentum for Nonconvex Optimization" by Kfir Yehuda Levy, Ali Kavis, Volkan Cevher
4. "Robust inverse Reinforcement Learning under Transition Dynamics Mismatch" by Luca Viano, Yu-Ting Huang, Parameswaran Kamalaruban, Adrian Weller, Volkan Cevher
5. "The Effect of the Intrinsic Dimension on the Generalization of Quadratic Classifiers" by Fabian Latorre, Leello Tadesse Dadi, Paul Rolland, Volkan Cevher
6. "Convergence of adaptive algorithms for constrained weakly convex optimization" by Ahmet Alacaoglu, Yura Malitsky, Volkan Cevher
7. "Subquadratic Overparameterization for Shallow Neural Networks" by Chaehwan Song, Ali Ramezani-Kebrya, Thomas Pethick, Armin Eftekhari, Volkan Cevher