Training Language GANs from Scratch
Cyprien de Masson d’Autume Mihaela Rosca Jack Rae Shakir Mohamed
DeepMind
{cyprien,mihaelacr,jwrae,shakir}@google.com
Abstract
Generative Adversarial Networks (GANs) enjoy great success at image genera-
tion, but have proven difficult to train in the domain of natural language. Chal-
lenges with gradient estimation, optimization instability, and mode collapse have
lea ...


雷达卡



京公网安备 11010802022788号







