Research map / Probabilistic and scientific modeling / Variational and generative sampling
InfoVAE: Balancing Learning and Inference in Variational Autoencoders
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 |
|---|---|---|---|
| replace ELBO with MMD-regularized InfoVAE objective | VAE · Binarized MNIST | log likelihood: 82.75 -> 80.76 | Helped |
| replace ELBO with Stein-regularized InfoVAE | VAE · Binarized MNIST | log likelihood: 82.75 -> 81.47 | Helped |
| replace ELBO with adversarial latent regularization | VAE · Binarized MNIST | log likelihood: 82.75 -> 82.21 | No clear effect |
| add large scaling lambda on q(z) divergence (InfoVAE) | VAE · MNIST small training sets | logdet covariance of q(z) | Helped |
| ELBO with PixelCNN decoder vs MMD regularization (alpha=1) | VAE · MNIST | semi-supervised classification error | Hurt |
| adversarial vs MMD latent regularization | VAE · MNIST | training speed/stability | Hurt |
| remove latent regularization (plain autoencoder) vs MMD (ablation) | VAE · MNIST | semi-supervised error | Mixed |
Closest papers by meaning
- Amortized Inference Regularization (NeurIPS 2018 · similarity 0.92)
- Variational Autoencoders with Jointly Optimized Latent Dependency Structure (ICLR 2019 · similarity 0.92)
- On Implicit Regularization in $β$-VAEs (ICML 2020 · similarity 0.92)
- Semi-Amortized Variational Autoencoders (ICML 2018 · similarity 0.92)
- Trading Information between Latents in Hierarchical Variational Autoencoders (ICLR 2023 · similarity 0.92)
- Variational Laplace Autoencoders (ICML 2019 · similarity 0.92)
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.