Omega: Optimistic EMA Gradients

Stochastic min-max optimization has gained interest in the machine learning community with the advancements in GANs and adversarial training. Although game optimization is fairly well understood in the deterministic setting, some issues persist in the stochastic regime. Recent work has shown that st...

Cijeli opis

Spremljeno u:
Bibliografski detalji
Glavni autori: Ramirez, Juan, Sukumaran, Rohan, Bertrand, Quentin, Gidel, Gauthier
Format: Članak
Izdano: 2023
Teme:
Online pristup:https://arxiv.org/abs/2306.07905
http://arxiv.org/abs/2306.07905
Oznake: Dodaj oznaku
Bez oznaka, Budi prvi tko označuje ovaj zapis!