# MHVAE: a Human-Inspired Deep Hierarchical Generative Model for Multimodal Representation Learning

arXiv: https://arxiv.org/abs/2006.02991 · HTML: https://mlsanity.com/paper/2006.02991/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jun 2020.

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

- [Learning Missing Modal Electronic Health Records with Unified Multi-modal Data Embedding and Modality-Aware Attention](https://mlsanity.com/paper/2305.02504/) (2023)
- [Generative AI for RF Sensing in IoT systems](https://mlsanity.com/paper/2407.07506/) (2024)
- [Composable Generative Models](https://mlsanity.com/paper/2102.09249/) (2021)
- [Benchmarking Multimodal Variational Autoencoders: CdSprites+ Dataset and Toolkit](https://mlsanity.com/paper/2209.03048/) (2022)
- [Bridging Language, Vision and Action: Multimodal VAEs in Robotic Manipulation Tasks](https://mlsanity.com/paper/2404.01932/) (2024)

## Abstract (arXiv)

Humans are able to create rich representations of their external reality. Their internal representations allow for cross-modality inference, where available perceptions can induce the perceptual experience of missing input modalities. In this paper, we contribute the Multimodal Hierarchical Variational Auto-encoder (MHVAE), a hierarchical multimodal generative model for representation learning. Inspired by human cognitive models, the MHVAE is able to learn modality-specific distributions, of an arbitrary number of modalities, and a joint-modality distribution, responsible for cross-modality inference. We formally derive the model's evidence lower bound and propose a novel methodology to approximate the joint-modality posterior based on modality-specific representation dropout. We evaluate the MHVAE on standard multimodal datasets. Our model performs on par with other state-of-the-art generative models regarding joint-modality reconstruction from arbitrary input modalities and cross-modality inference.