# Learning Deep Energy Models: Contrastive Divergence vs. Amortized MLE

arXiv: https://arxiv.org/abs/1707.00797 · HTML: https://mlsanity.com/paper/1707.00797/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jul 2017.

## Built on by (most cited first, 5 of 5)

- [Layer Collaboration in the Forward-Forward Algorithm](https://mlsanity.com/paper/2305.12393/) (2023)
- [Learning Equivariant Energy Based Models with Equivariant Stein Variational Gradient Descent](https://mlsanity.com/paper/2106.07832/) (2021)
- [Training Energy-Based Models with Diffusion Contrastive Divergences](https://mlsanity.com/paper/2307.01668/) (2023)
- [Score Matching With Missing Data](https://mlsanity.com/paper/2506.00557/) (2025)
- [AmorLIP: Efficient Language-Image Pretraining via Amortization](https://mlsanity.com/paper/2505.18983/) (2025)

## Abstract (arXiv)

We propose a number of new algorithms for learning deep energy models and demonstrate their properties. We show that our SteinCD performs well in term of test likelihood, while SteinGAN performs well in terms of generating realistic looking images. Our results suggest promising directions for learning better models by combining GAN-style methods with traditional energy-based learning.