§ feed · storyline
Implicit generation and generalization methods for energy-based models
OpenAI publishes research on energy-based models showing improved sample quality and generalisation via stable training methods that rival GANs while maintaining likelihood-based mode coverage.
We’ve made progress towards stable and scalable training of energy-based models (EBMs) resulting in better sample quality and generalization ability than existing models. Generation in EBMs spends more compute to continually refine its answers and doing so can generate samples competitive with GANs at low temperatures, while also having mode coverage guarantees of likelihood-based models.
We hope these findings stimulate further research into this promising class of models.
§ sources1 publication · timeline below