Research map / Probabilistic and scientific modeling / Variational and generative sampling
Learning Autoencoders with Relational Regularization
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 relational regularization with a learnable structured prior in a probabilistic autoencoder | P-RAE · MNIST | FID: 41.99 vs 35.63 | Hurt |
| Use relational regularization with a learnable structured prior in a probabilistic autoencoder | P-RAE · MNIST test | reconstruction loss: 16.60 -> 14.14 | Helped |
| Use relational regularization with a learnable structured prior in a probabilistic autoencoder | P-RAE · CelebA | FID: 52.07 vs 52.20 | Hurt |
| Use relational regularization with a learnable structured prior in a probabilistic autoencoder | P-RAE · CelebA test | reconstruction loss: 63.83 -> 63.21 | Helped |
| Use relational regularization with a learnable structured prior in a deterministic autoencoder | D-RAE · MNIST | FID: 35.63 vs 49.39 | Hurt |
| Use relational regularization with a learnable structured prior in a deterministic autoencoder | D-RAE · MNIST test | reconstruction loss: 9.97 -> 10.37 | Hurt |
| Use relational regularization with a learnable structured prior in a deterministic autoencoder | D-RAE · CelebA | FID: 52.07 -> 51.45 | Helped |
| Use relational regularization with a learnable structured prior in a deterministic autoencoder | D-RAE · CelebA test | reconstruction loss: 63.83 -> 64.49 | Hurt |
| Replace Wasserstein regularization with relational GW regularization for co-training | AEs+GW · Caltech101-7 | classification accuracy: 83.07±1.69 -> 84.29±1.74 | Helped |
| Replace Wasserstein regularization with relational GW regularization for co-training | AEs+GW · Caltech101-20 | classification accuracy: 69.58±2.03 -> 69.39±2.01 | No clear effect |
| Replace Wasserstein regularization with relational GW regularization for co-training | AEs+GW · Handwritten | classification accuracy: 71.21±5.55 -> 72.36±3.82 | Helped |
| Replace Wasserstein regularization with relational GW regularization for co-training | AEs+GW · Cathgen | classification accuracy: 66.06±1.68 -> 66.99±1.77 | Helped |
Closest papers by meaning
- Improving Relational Regularized Autoencoders with Spherical Sliced Fused Gromov Wasserstein (ICLR 2021 · similarity 0.92)
- Shape your Space: A Gaussian Mixture Regularization Approach to Deterministic Autoencoders (NeurIPS 2021 · similarity 0.89)
- MAE: Mutual Posterior-Divergence Regularization for Variational AutoEncoders (ICLR 2019 · similarity 0.88)
- A Statistical Analysis of Wasserstein Autoencoders for Intrinsically Low-dimensional Data (ICLR 2024 · similarity 0.87)
- Sparse Autoencoders, Again? (ICML 2025 · similarity 0.86)
- Masked Autoencoders Are Effective Tokenizers for Diffusion Models (ICML 2025 · similarity 0.86)
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.