# Learning Latent Subspaces in Variational Autoencoders

arXiv: https://arxiv.org/abs/1812.06190 · HTML: https://mlsanity.com/paper/1812.06190/

## Key facts

- Cited by 51 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Dec 2018.

## Built on by (most cited first, 30 of 51)

- [Promises and Pitfalls of Black-Box Concept Learning Models](https://mlsanity.com/paper/2106.13314/) (2021)
- [Identifiability of deep generative models without auxiliary information](https://mlsanity.com/paper/2206.10044/) (2022)
- [DISSECT: Disentangled Simultaneous Explanations via Concept Traversals](https://mlsanity.com/paper/2105.15164/) (2021)
- [Generalizing Nonlinear ICA Beyond Structural Sparsity](https://mlsanity.com/paper/2311.00866/) (2023)
- [FaceController: Controllable Attribute Editing for Face in the Wild](https://mlsanity.com/paper/2102.11464/) (2021)
- [Generating Features with Increased Crop-related Diversity for Few-Shot Object Detection](https://mlsanity.com/paper/2304.05096/) (2023)
- [Efficient Out-of-Distribution Detection Using Latent Space of β-VAE for Cyber-Physical Systems](https://mlsanity.com/paper/2108.11800/) (2021)
- [Lifelong Teacher-Student Network Learning](https://mlsanity.com/paper/2107.04689/) (2021)
- [Target-aware Molecular Graph Generation](https://mlsanity.com/paper/2202.04829/) (2022)
- [Decoupling Local and Global Representations of Time Series](https://mlsanity.com/paper/2202.02262/) (2022)
- [PROUD: PaRetO-gUided Diffusion Model for Multi-objective Generation](https://mlsanity.com/paper/2407.04493/) (2024)
- [Comparing the latent space of generative models](https://mlsanity.com/paper/2207.06812/) (2022)
- [Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations](https://mlsanity.com/paper/2209.13446/) (2022)
- [Multi-objective Deep Data Generation with Correlated Property Control](https://mlsanity.com/paper/2210.01796/) (2022)
- [Exploring Behavior-Relevant and Disentangled Neural Dynamics with Generative Diffusion Models](https://mlsanity.com/paper/2410.09614/) (2024)
- [Deep Image Synthesis from Intuitive User Input: A Review and Perspectives](https://mlsanity.com/paper/2107.04240/) (2021)
- [Face Identity-Aware Disentanglement in StyleGAN](https://mlsanity.com/paper/2309.12033/) (2023)
- [Disentangling representations of retinal images with generative models](https://mlsanity.com/paper/2402.19186/) (2024)
- [Semi-Supervised Disentanglement of Class-Related and Class-Independent Factors in VAE](https://mlsanity.com/paper/2102.00892/) (2021)
- [A multimodal dynamical variational autoencoder for audiovisual speech representation learning](https://mlsanity.com/paper/2305.03582/) (2023)
- [PluGeN: Multi-Label Conditional Generation From Pre-Trained Models](https://mlsanity.com/paper/2109.09011/) (2021)
- [Learning Invariances with Generalised Input-Convex Neural Networks](https://mlsanity.com/paper/2204.07009/) (2022)
- [Learning Robust Latent Representations for Controllable Speech Synthesis](https://mlsanity.com/paper/2105.04458/) (2021)
- [Learning Conditional Invariance through Cycle Consistency](https://mlsanity.com/paper/2111.13185/) (2021)
- [Gradient-based Counterfactual Explanations using Tractable Probabilistic Models](https://mlsanity.com/paper/2205.07774/) (2022)
- [Outcome-Guided Counterfactuals for Reinforcement Learning Agents from a Jointly Trained Generative Latent Space](https://mlsanity.com/paper/2207.07710/) (2022)
- [Predefined Prototypes for Intra-Class Separation and Disentanglement](https://mlsanity.com/paper/2406.16145/) (2024)
- [Viscos Flows: Variational Schur Conditional Sampling With Normalizing Flows](https://mlsanity.com/paper/2107.02474/) (2021)
- [Invariance-based Multi-Clustering of Latent Space Embeddings for Equivariant Learning](https://mlsanity.com/paper/2107.11717/) (2021)
- [Group-disentangled Representation Learning with Weakly-Supervised Regularization](https://mlsanity.com/paper/2110.12185/) (2021)

## Abstract (arXiv)

Variational autoencoders (VAEs) are widely used deep generative models capable of learning unsupervised latent representations of data. Such representations are often difficult to interpret or control. We consider the problem of unsupervised learning of features correlated to specific labels in a dataset. We propose a VAE-based generative model which we show is capable of extracting features correlated to binary labels in the data and structuring it in a latent subspace which is easy to interpret. Our model, the Conditional Subspace VAE (CSVAE), uses mutual information minimization to learn a low-dimensional latent subspace associated with each label that can easily be inspected and independently manipulated. We demonstrate the utility of the learned representations for attribute manipulation tasks on both the Toronto Face and CelebA datasets.