Research map / Probabilistic and scientific modeling / Variational and generative sampling
Nonparametric Variational Auto-Encoders for Hierarchical Representation Learning
Prasoon Goyal, Zhiting Hu, Xiaodan Liang, Chenyu Wang, Eric P. Xing
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace standard normal prior with a learned nCRP tree prior | VAE · TRECVID MED 2011 test frames | mean test log-likelihood: -28886.90 -> -28438.32 | Improved |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 (aggregate) | classification accuracy (%): 39.1 -> 42.4 | Improved |
| replace K-Means clustering with VAE-nCRP | frame features · TRECVID MED 2011 (aggregate) | classification accuracy (%): 34.9 -> 42.4 | Improved |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 (aggregate) | retrieval F-1 (%): 38.5 -> 42.4 | Improved |
| replace K-Means clustering with VAE-nCRP | frame features · TRECVID MED 2011 (aggregate) | retrieval F-1 (%): 32.4 -> 42.4 | Improved |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 - Birthday_party | classification accuracy (%): 7.4 -> 27.8 | Improved |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 - Woodworking | classification accuracy (%): 40.5 -> 60.8 | Improved |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 - Board_trick | classification accuracy (%): 47.2 -> 31.3 | Hurt |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 - Parade | classification accuracy (%): 33.8 -> 19.8 | Hurt |
| replace Gaussian mixture prior with nCRP prior | VAE · TRECVID MED 2011 - 15 classes | classification accuracy (%): 8 of 15 classes better | Mixed |
Closest papers by meaning
- Trading Information between Latents in Hierarchical Variational Autoencoders (ICLR 2023 · similarity 0.92)
- A Bayesian Nonparametrics View into Deep Representations (NeurIPS 2020 · similarity 0.92)
- Learning Hierarchical Priors in VAEs (NeurIPS 2019 · similarity 0.91)
- Consistency Regularization for Variational Auto-Encoders (NeurIPS 2021 · similarity 0.91)
- Beyond Vanilla Variational Autoencoders: Detecting Posterior Collapse in Conditional and Hierarchical Variational Autoencoders (ICLR 2024 · similarity 0.90)
- Variational Learning of Fractional Posteriors (ICML 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
