Corpus ID: 223953581

Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models

@article{Bao2020VariationalE,
  title={Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models},
  author={Fan Bao and K. Xu and Chongxuan Li and Lanqing Hong and J. Zhu and Bo Zhang},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.08258}
}
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… Expand

Figures from this paper

References

SHOWING 1-10 OF 38 REFERENCES
Bi-level Score Matching for Learning Energy-based Latent Variable Models
  • 3
  • PDF
Variational Noise-Contrastive Estimation
  • 8
  • PDF
Learning Doubly Intractable Latent Variable Models via Score Matching
  • 2
  • PDF
Flow Contrastive Estimation of Energy-Based Models
  • 28
  • PDF
A Tutorial on Energy-Based Learning
  • 597
  • Highly Influential
  • PDF
On Autoencoders and Score Matching for Energy Based Models
  • 80
  • PDF
Implicit Generation and Generalization in Energy-Based Models
  • 89
  • PDF
...
1
2
3
4
...