mlsanityData to 7 Oct 2026

Learning Latent Subspaces in Variational Autoencoders

arXiv 1812.06190 ↗First posted 14 Dec 2018cs.LG cs.CV stat.MLcited by 51

Not on the map. Why?

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.

Key facts. Cited by 51 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Dec 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 17; 2022: 11; 2023: 6; 2024: 9; 2025: 4; 2026: 4.

51 papers.

Sort:

By area:

  1. 2021Promises and Pitfalls of Black-Box Concept Learning Modelscited by 128
  2. 2022Identifiability of deep generative models without auxiliary informationcited by 70
  3. 2021DISSECT: Disentangled Simultaneous Explanations via Concept Traversalscited by 37
  4. 2023Generalizing Nonlinear ICA Beyond Structural Sparsitycited by 33
  5. 2021FaceController: Controllable Attribute Editing for Face in the Wildcited by 22
  6. 2023Generating Features with Increased Crop-related Diversity for Few-Shot Object Detectioncited by 21
  7. 2021Efficient Out-of-Distribution Detection Using Latent Space of β-VAE for Cyber-Physical Systemscited by 19
  8. 2021Lifelong Teacher-Student Network Learningcited by 18
  9. 2022Target-aware Molecular Graph Generationcited by 18
  10. 2022Decoupling Local and Global Representations of Time Seriescited by 17
  11. 2024PROUD: PaRetO-gUided Diffusion Model for Multi-objective Generationcited by 15
  12. 2022Comparing the latent space of generative modelscited by 11
38 more
  1. 2022Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanationscited by 11
  2. 2022Multi-objective Deep Data Generation with Correlated Property Controlcited by 10
  3. 2024Exploring Behavior-Relevant and Disentangled Neural Dynamics with Generative Diffusion Modelscited by 9
  4. 2021Deep Image Synthesis from Intuitive User Input: A Review and Perspectivescited by 8
  5. 2023Face Identity-Aware Disentanglement in StyleGANcited by 8
  6. 2024Disentangling representations of retinal images with generative modelscited by 8
  7. 2021Semi-Supervised Disentanglement of Class-Related and Class-Independent Factors in VAEcited by 7
  8. 2023A multimodal dynamical variational autoencoder for audiovisual speech representation learningcited by 6
  9. 2021PluGeN: Multi-Label Conditional Generation From Pre-Trained Modelscited by 5
  10. 2022Learning Invariances with Generalised Input-Convex Neural Networkscited by 3
  11. 2021Learning Robust Latent Representations for Controllable Speech Synthesiscited by 2
  12. 2021Learning Conditional Invariance through Cycle Consistencycited by 2
  13. 2022Gradient-based Counterfactual Explanations using Tractable Probabilistic Modelscited by 2
  14. 2022Outcome-Guided Counterfactuals for Reinforcement Learning Agents from a Jointly Trained Generative Latent Spacecited by 2
  15. 2024Predefined Prototypes for Intra-Class Separation and Disentanglementcited by 2
  16. 2021Viscos Flows: Variational Schur Conditional Sampling With Normalizing Flowscited by 1
  17. 2021Invariance-based Multi-Clustering of Latent Space Embeddings for Equivariant Learningcited by 1
  18. 2021Group-disentangled Representation Learning with Weakly-Supervised Regularizationcited by 1
  19. 2022Repairing Systematic Outliers by Learning Clean Subspaces in VAEscited by 1
  20. 2023Variational Autoencoders for Feature Exploration and Malignancy Prediction of Lung Lesionscited by 1
  21. 2024Exploring Text-Guided Single Image Editing for Remote Sensing Imagescited by 1
  22. 2024Particle physics DL-simulation with control over generated data propertiescited by 1
  23. 2024Structural Disentanglement of Causal and Correlated Conceptscited by 1
  24. 2024LatentExplainer: Explaining Latent Representations in Deep Generative Models with Multimodal Large Language Modelscited by 1
  25. 2025Variational Learning of Disentangled Representationscited by 1
  26. 2025Explainable Disentanglement on Discrete Speech Representations for Noise-Robust ASRcited by 1
  27. 2021Advances in Electron Microscopy with Deep Learningcited by 0
  28. 2021Deep Contrastive Patch-Based Subspace Learning for Camera Image Signal Processingcited by 0
  29. 2021Proceedings of the First Workshop on Weakly Supervised Learning (WeaSuL)cited by 0
  30. 2021Redatuming physical systems using symmetric autoencoderscited by 0
  31. 2022Provable Clustering of a Union of Linear Manifolds Using Optimal Directionscited by 0
  32. 2023Controllable Data Generation Via Iterative Data-Property Mutual Mappingscited by 0
  33. 2024Unsupervised Panoptic Interpretation of Latent Spaces in GANs Using Space-Filling Vector Quantizationcited by 0
  34. 2025GPO-VAE: Modeling Explainable Gene Perturbation Responses utilizing GRN-Aligned Parameter Optimizationcited by 0
  35. 2025Video Anomaly Detection with Contours -- A Studycited by 0
  36. 2026Distributed Deep Variational Approach for Privacy-preserving Data Releasecited by 0
  37. 2026Environment-Robust Representation Learning with Empirical Bayescited by 0
  38. 2026The Trade-off Between Covariate Dependence and Latent Structure in Representation Learningcited by 0

50 most cited of 51 All 51 (JSON)