Research map / Probabilistic and scientific modeling / Variational and generative sampling
Learning Latent Superstructures in Variational Autoencoders for Deep Multidimensional Clustering
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use LTVAE instead of IWAE | VAE generative modeling · MNIST test | test loglikelihood: -85.39$\pm$0.13 -> -84.75$\pm$0.14 | Helped |
| use LTVAE instead of IWAE | VAE generative modeling · STL test | test loglikelihood: -742.43$\pm$1.10 -> -619.04$\pm$7.88 | Helped |
| use LTVAE instead of IWAE | VAE generative modeling · Reuters test | test loglikelihood: -1254.29$\pm$6.95 -> -1245.71$\pm$4.45 | Helped |
| use LTVAE instead of IWAE | VAE generative modeling · HHAR test | test loglikelihood: -16.53$\pm$0.16 -> -13.65$\pm$0.53 | Helped |
| use LTVAE instead of DCN | deep clustering · MNIST | clustering accuracy: 83.32\% -> 86.32\% | Helped |
| use LTVAE instead of DCN | deep clustering · STL-10 | clustering accuracy: 85.88\% -> 90.00\% | Helped |
| use LTVAE instead of DCN | deep clustering · Reuters | clustering accuracy: 75.05\% -> 80.96\% | Helped |
| use LTVAE instead of DCN | deep clustering · HHAR | clustering accuracy: 81.26\% -> 85.00\% | Helped |
| fix the latent tree to one latent variable connected to all z nodes (ablation) | LTVAE · MNIST | computational time (s): 7,592 -> 3,197 | Helped |
| fix the latent tree to one latent variable connected to all z nodes (ablation) | LTVAE · STL-10 | computational time (s): 3,251 -> 542 | Helped |
| fix the latent tree to one latent variable connected to all z nodes (ablation) | LTVAE · Reuters | computational time (s): 5,756 -> 1,442 | Helped |
| fix the latent tree to one latent variable connected to all z nodes (ablation) | LTVAE · HHAR | computational time (s): 5,693 -> 1,021 | Helped |
Closest papers by meaning
- Multi-Facet Clustering Variational Autoencoders (NeurIPS 2021 · similarity 0.92)
- Tree Variational Autoencoders (NeurIPS 2023 · similarity 0.91)
- Variational Autoencoders with Jointly Optimized Latent Dependency Structure (ICLR 2019 · similarity 0.90)
- Shape your Space: A Gaussian Mixture Regularization Approach to Deterministic Autoencoders (NeurIPS 2021 · similarity 0.89)
- Disentangling Disentanglement in Variational Autoencoders (ICML 2019 · similarity 0.88)
- $t^3$-Variational Autoencoder: Learning Heavy-tailed Data with Student's t and Power Divergence (ICLR 2024 · similarity 0.87)
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.