# Multimodal hierarchical Variational AutoEncoders with Factor Analysis latent space

arXiv: https://arxiv.org/abs/2207.09185 · HTML: https://mlsanity.com/paper/2207.09185/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 42 references that are in the map.
- First posted on arXiv 19 Jul 2022.

## Builds on (most cited first, 21 of 21)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Variational Inference: A Review for Statisticians](https://mlsanity.com/paper/1601.00670/) (2016)
- [Understanding disentangling in β-VAE](https://mlsanity.com/paper/1804.03599/) (2018)
- [Ladder Variational Autoencoders](https://mlsanity.com/paper/1602.02282/) (2016)
- [Wav2CLIP: Learning Robust Audio Representations From CLIP](https://mlsanity.com/paper/2110.11499/) (2021)
- [Multimodal Generative Models for Scalable Weakly-Supervised Learning](https://mlsanity.com/paper/1802.05335/) (2018)
- [Variational Mixture-of-Experts Autoencoders for Multi-Modal Deep Generative Models](https://mlsanity.com/paper/1911.03393/) (2019)
- [ADD 2022: the First Audio Deep Synthesis Detection Challenge](https://mlsanity.com/paper/2202.08433/) (2022)
- [Joint Multimodal Learning with Deep Generative Models](https://mlsanity.com/paper/1611.01891/) (2016)
- [Learning Hierarchical Priors in VAEs](https://mlsanity.com/paper/1905.04982/) (2019)
- [Multi-VAE: Learning Disentangled View-common and View-peculiar Visual Representations for Multi-view Clustering](https://mlsanity.com/paper/2106.11232/) (2021)
- [Hamiltonian Variational Auto-Encoder](https://mlsanity.com/paper/1805.11328/) (2018)
- [XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings](https://mlsanity.com/paper/1711.05139/) (2017)
- [Missing Data Imputation and Acquisition with Deep Hierarchical Models and Hamiltonian Monte Carlo](https://mlsanity.com/paper/2202.04599/) (2022)
- [NestedVAE: Isolating Common Factors via Weak Supervision](https://mlsanity.com/paper/2002.11576/) (2020)
- [Sparse Semi-supervised Heterogeneous Interbattery Bayesian Analysis](https://mlsanity.com/paper/2001.08975/) (2020)
- [Unsupervised Learning of Global Factors in Deep Generative Models](https://mlsanity.com/paper/2012.08234/) (2020)
- [Organization of a Latent Space structure in VAE/GAN trained by navigation data](https://mlsanity.com/paper/2102.01852/) (2021)
- [Bayesian Sparse Factor Analysis with Kernelized Observations](https://mlsanity.com/paper/2006.00968/) (2020)
- [Table Caption Generation in Scholarly Documents Leveraging Pre-trained Language Models](https://mlsanity.com/paper/2108.08111/) (2021)

## Built on by (most cited first, 2 of 2)

- [Disentangling Genotype and Environment Specific Latent Features for Improved Trait Prediction using a Compositional Autoencoder](https://mlsanity.com/paper/2410.19922/) (2024)
- [MediEncoder: Nonlinear Representation Learning for High-Dimensional Causal Mediation Analysis](https://mlsanity.com/paper/2606.30648/) (2026)

## Abstract (arXiv)

Purpose: Handling heterogeneous and mixed data types has become increasingly critical with the exponential growth in real-world databases. While deep generative models attempt to merge diverse data views into a common latent space, they often sacrifice interpretability, flexibility, and modularity. This study proposes a novel method to address these limitations by combining Variational AutoEncoders (VAEs) with a Factor Analysis latent space (FA-VAE). Methods: The proposed FA-VAE method employs multiple VAEs to learn a private representation for each heterogeneous data view in a continuous latent space. Information is shared between views using a low-dimensional latent space, generated via a linear projection matrix. This modular design creates a hierarchical dependency between private and shared latent spaces, allowing for the flexible addition of new views and conditioning of pre-trained models. Results: The FA-VAE approach facilitates cross-generation of data from different domains and enables transfer learning between generative models. This allows for effective integration of information across diverse data views while preserving their distinct characteristics. Conclusions: By overcoming the limitations of existing methods, the FA-VAE provides a more interpretable, flexible, and modular solution for managing heterogeneous data types. It offers a pathway to more efficient and scalable data-handling strategies, enhancing the potential for cross-domain data synthesis and model transferability.