OpenAI Blog

Implicit generation and generalization methods for energy-based models

Read the original on OpenAI Blog →

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.

Summary generated by The Flow from the publisher's feed. The full article lives at OpenAI Blog.

Hugging Face Trending Papers
Jul 29

Equilibrium Training of Energy-Based Models with Parallel Trajectory Tempering

Energy-Based Models (EBMs) provide an interpretable framework for generative modeling of scientific data, but poor Markov Chain Monte Carlo mixing often limits their reliability. We introduce a training algorithm based on Parallel Trajectory Tempering (PTT), which exploits the continuity of the optimization path to maintain equilibrium sampling throughout learning.

arXiv AI
Jun 24

Catastrophic Compositional Generation: Why Vanilla Diffusion Models Fail to Extrapolate

arXiv:2606. 23920v1 Announce Type: cross Abstract: The task of compositional generation involves using a conditional generative model, trained only on a subset of the possible conditions, to produce samples from compositionally-defined target distributions such as a geometric combination of the source distributions.

By Duncan Soiffer, Chandler Squires, Yuan Guan, Jason Hartford, Pradeep Ravikumar
arXiv Machine Learning
Jul 2

Decision-Aware Training for Sample-Based Generative Models

arXiv:2607. 01171v1 Announce Type: new Abstract: Sample-based generative models are increasingly used for probabilistic forecasting in high-stakes decision settings, yet their training objectives are blind to the decision maker's cost structure.

By Kornelius Raeth, Nicole Ludwig