Timezone: »
The learning and evaluation of energy-based latent variable models (EBLVMs) without any structural assumptions are highly challenging, because the true posteriors and the partition functions in such models are generally intractable. This paper presents variational estimates of the score function and its gradient with respect to the model parameters in a general EBLVM, referred to as VaES and VaGES respectively. The variational posterior is trained to minimize a certain divergence to the true model posterior and the bias in both estimates can be bounded by the divergence theoretically. With a minimal model assumption, VaES and VaGES can be applied to the kernelized Stein discrepancy (KSD) and score matching (SM)-based methods to learn EBLVMs. Besides, VaES can also be used to estimate the exact Fisher divergence between the data and general EBLVMs.
Author Information
Fan Bao (Tsinghua University)
More from the Same Authors
-
2022 Poster: EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations »
Min Zhao · Fan Bao · Chongxuan LI · Jun Zhu -
2022 Poster: DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps »
Cheng Lu · Yuhao Zhou · Fan Bao · Jianfei Chen · Chongxuan LI · Jun Zhu -
2022 : All are Worth Words: a ViT Backbone for Score-based Diffusion Models »
Fan Bao · Chongxuan LI · Yue Cao · Jun Zhu -
2022 : Why Are Conditional Generative Models Better Than Unconditional Ones? »
Fan Bao · Chongxuan LI · Jiacheng Sun · Jun Zhu -
2022 Spotlight: EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations »
Min Zhao · Fan Bao · Chongxuan LI · Jun Zhu -
2022 Spotlight: DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps »
Cheng Lu · Yuhao Zhou · Fan Bao · Jianfei Chen · Chongxuan LI · Jun Zhu -
2021 Poster: Stability and Generalization of Bilevel Programming in Hyperparameter Optimization »
Fan Bao · Guoqiang Wu · Chongxuan LI · Jun Zhu · Bo Zhang -
2020 Poster: Bi-level Score Matching for Learning Energy-based Latent Variable Models »
Fan Bao · Chongxuan LI · Kun Xu · Hang Su · Jun Zhu · Bo Zhang