mlsanityData to 7 Oct 2026

Learning Hierarchical Features from Generative Models

arXiv 1702.08396 ↗First posted 27 Feb 2017cs.LG stat.MLcited by 27

Not on the map. Why?

Abstract (arXiv)

Deep neural networks have been shown to be very successful at learning feature hierarchies in supervised learning tasks. Generative models, on the other hand, have benefited less from hierarchical models with multiple layers of latent variables. In this paper, we prove that hierarchical latent variable models do not take advantage of the hierarchical structure when trained with existing variational methods, and provide some limitations on the kind of features existing models can learn. Finally we propose an alternative architecture that do not suffer from these limitations. Our model is able to learn highly interpretable and disentangled hierarchical features on several natural image datasets with no task specific regularization or prior knowledge.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Feb 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 10; 2022: 3; 2023: 6; 2024: 3; 2025: 4; 2026: 1.

27 papers.

Sort:

By area:

  1. 2022Deep Clustering: A Comprehensive Surveycited by 78
  2. 2021Greedy Hierarchical Variational Autoencoders for Large-Scale Video Predictioncited by 75
  3. 2021Clockwork Variational Autoencoderscited by 40
  4. 2023Understanding Masked Autoencoders via Hierarchical Latent Variable Modelscited by 39
  5. 2021Diagonal Attention and Style-based GAN for Content-Style Disentanglement in Image Generation and Translationcited by 35
  6. 2021Where and What? Examining Interpretable Disentangled Representationscited by 27
  7. 2021Commutative Lie Group VAE for Disentanglement Learningcited by 27
  8. 2023Multifactor Sequential Disentanglement via Structured Koopman Autoencoderscited by 23
  9. 2021Learning Interpretable Representations of Entanglement in Quantum Optics Experiments using Deep Generative Modelscited by 12
  10. 2021On Incorporating Inductive Biases into VAEscited by 11
  11. 2025Scale-wise Distillation of Diffusion Modelscited by 11
  12. 2021Unsupervised Learning of Neurosymbolic Encoderscited by 8
15 more
  1. 2023Learning Joint Latent Space EBM Prior Model for Multi-layer Generatorcited by 8
  2. 2021Multi-Task Neural Processescited by 7
  3. 2022Gromov-Wasserstein Autoencoderscited by 7
  4. 2023Beyond Vanilla Variational Autoencoders: Detecting Posterior Collapse in Conditional and Hierarchical Variational Autoencoderscited by 6
  5. 2023Learning Hierarchical Features with Joint Latent Space Energy-Based Priorcited by 6
  6. 2024Learning Latent Space Hierarchical EBM Diffusion Modelscited by 6
  7. 2024Nested Diffusion Models Using Hierarchical Latent Priorscited by 2
  8. 2023Computing with Categories in Machine Learningcited by 1
  9. 2025Evolved Hierarchical Masking for Self-Supervised Learningcited by 1
  10. 2021Meta-learning Amidst Heterogeneity and Ambiguitycited by 0
  11. 2022A Probabilistic Generative Model of Free Categoriescited by 0
  12. 2024CAVACHON: a hierarchical variational autoencoder to integrate multi-modal single-cell datacited by 0
  13. 2025Likelihood-Free Variational Autoencoderscited by 0
  14. 2025Contrastive Deep Learning for Variant Detection in Wastewater Genomic Sequencingcited by 0
  15. 2026Latent Fourier Transformcited by 0