Balanced Training of Energy-Based Models with Adaptive Flow Sampling

arXiv:2306.00684v4 Announce Type: replace
Abstract: Energy-based models (EBMs) are versatile density estimation models that directly parameterize an unnormalized log density. Although very flexible, EBMs lack a specified normalization constant of the model, making the likelihood of the model computationally intractable. Several approximate samplers and variational inference techniques have been proposed to estimate the likelihood gradients for training. These techniques have shown promising results in generating samples, but little attention has been paid to the statistical accuracy of the estimated density, …

abstract arxiv cs.lg energy flow inference likelihood making normalization sampling stat.ml training type