# Learning Representations by Maximizing Mutual Information in Variational Autoencoders

arXiv: https://arxiv.org/abs/1912.13361 · HTML: https://mlsanity.com/paper/1912.13361/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Dec 2019.

## Built on by (most cited first, 20 of 20)

- [Diffusion Model with Cross Attention as an Inductive Bias for Disentanglement](https://mlsanity.com/paper/2402.09712/) (2024)
- [Behavior-Driven Synthesis of Human Dynamics](https://mlsanity.com/paper/2103.04677/) (2021)
- [Internally Rewarded Reinforcement Learning](https://mlsanity.com/paper/2302.00270/) (2023)
- [Mutual Information-calibrated Conformal Feature Fusion for Uncertainty-Aware Multimodal 3D Object Detection at the Edge](https://mlsanity.com/paper/2309.09593/) (2023)
- [Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness](https://mlsanity.com/paper/2110.12381/) (2021)
- [Forget-me-not! Contrastive Critics for Mitigating Posterior Collapse](https://mlsanity.com/paper/2207.09535/) (2022)
- [Discriminative Multimodal Learning via Conditional Priors in Generative Models](https://mlsanity.com/paper/2110.04616/) (2021)
- [Multi-Modal Mutual Information Maximization: A Novel Approach for Unsupervised Deep Cross-Modal Hashing](https://mlsanity.com/paper/2112.06489/) (2021)
- [PCAE: A Framework of Plug-in Conditional Auto-Encoder for Controllable Text Generation](https://mlsanity.com/paper/2210.03496/) (2022)
- [CR-VAE: Contrastive Regularization on Variational Autoencoders for Preventing Posterior Collapse](https://mlsanity.com/paper/2309.02968/) (2023)
- [DynaSeg: A Deep Dynamic Fusion Method for Unsupervised Image Segmentation Incorporating Feature Similarity and Spatial Continuity](https://mlsanity.com/paper/2405.05477/) (2024)
- [Training β-VAE by Aggregating a Learned Gaussian Posterior with a Decoupled Decoder](https://mlsanity.com/paper/2209.14783/) (2022)
- [Causally Disentangled Generative Variational AutoEncoder](https://mlsanity.com/paper/2302.11737/) (2023)
- [VOLTA: Improving Generative Diversity by Variational Mutual Information Maximizing Autoencoder](https://mlsanity.com/paper/2307.00852/) (2023)
- [Semantic-Based Few-Shot Learning by Interactive Psychometric Testing](https://mlsanity.com/paper/2112.09201/) (2021)
- [How to Construct Energy for Images? Denoising Autoencoder Can Be Energy Based Model](https://mlsanity.com/paper/2303.03887/) (2023)
- [Sample Observed Effects: Enumeration, Randomization and Generalization](https://mlsanity.com/paper/2108.04376/) (2021)
- [Deep Dive into Semi-Supervised ELBO for Improving Classification Performance](https://mlsanity.com/paper/2108.12734/) (2021)
- [Tracking Object Positions in Reinforcement Learning: A Metric for Keypoint Detection (extended version)](https://mlsanity.com/paper/2312.00592/) (2023)
- [SPIBER: Reconstructing Free Energy Landscapes from Short, Unconverged Trajectories with Generative Flow Networks](https://mlsanity.com/paper/2609.22663/) (2026)

## Abstract (arXiv)

Variational autoencoders (VAEs) have ushered in a new era of unsupervised learning methods for complex distributions. Although these techniques are elegant in their approach, they are typically not useful for representation learning. In this work, we propose a simple yet powerful class of VAEs that simultaneously result in meaningful learned representations. Our solution is to combine traditional VAEs with mutual information maximization, with the goal to enhance amortized inference in VAEs using Information Theoretic techniques. We call this approach InfoMax-VAE, and such an approach can significantly boost the quality of learned high-level representations. We realize this through the explicit maximization of information measures associated with the representation. Using extensive experiments on varied datasets and setups, we show that InfoMax-VAE outperforms contemporary popular approaches, including Info-VAE and β-VAE.