Session WB-1: Advances in stochastic and non-euclidean first order methods in stream Zeroth and first-order optimization methods
Wednesday, 10:30-12:30Room: B100/1001
| Session chair(s): |
|
| 222. Non-monotone stochastic line search without overhead for training neural networks |
Andrea Cristofari
[R] - Italy | accepted | ||
| Leonardo Galli
[R] - Germany | ||||
| Stefano Lucidi
[] - Italy | ||||
| 223. On the convergence of stochastic Bregman proximal gradient algorithms with biased gradient estimators |
Thomas Guilmeau
[R] - France | accepted | ||
| Emilie Chouzenoux
[] - France | ||||
| Víctor Elvira
[] - United Kingdom | ||||
| 224. Riemannian gradient descent improves parameter-efficient fine-tuning |
Bingcong Li
[R] - Switzerland | accepted | ||
| 225. General Tail Bounds for Non-Smooth Stochastic Mirror Descent |
Andrea Paudice
[R] - Denmark | accepted | ||
| Khaled Eldowa
[] - Italy | ||||