Research map / Label-efficient and robust learning / Recognition and long-tail
Disentangling Factors of Variation by Mixing Them
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 |
|---|---|---|---|
| add mixing autoencoder to a vanilla autoencoder (ablation) | Sprites feature learning · Sprites | mean average precision (average): 0.46 -> 0.62 | Helped |
| add both classifier and adversarial losses to the mixing autoencoder (ablation) | MIX · Sprites | mean average precision (average): 0.62 -> 0.70 | Helped |
| add classifier loss without adversarial loss (ablation) | MIX · Sprites | mean average precision (average): 0.62 -> 0.56 | Hurt |
| add adversarial loss without classifier loss (ablation) | MIX · Sprites | mean average precision (average): 0.62 -> 0.51 | Hurt |
| use BEGAN rather than DCGAN for adversarial training (ablation) | the proposed method · CelebA | classification performance (Avg.): 79.5 -> 81.7 | Helped |
| increase β in β-VAE from 2 to 8 (ablation) | β-VAE · CelebA | classification performance (Avg.): 81.3 -> 79.2 | Hurt |
Closest papers by meaning
- An Image is Worth More Than a Thousand Words: Towards Disentanglement in The Wild (NeurIPS 2021 · similarity 0.90)
- Disentangled Representation Learning via Modular Compositional Bias (NeurIPS 2025 · similarity 0.89)
- Disentangling by Factorising (ICML 2018 · similarity 0.89)
- Overcoming the Disentanglement vs Reconstruction Trade-off via Jacobian Supervision (ICLR 2019 · similarity 0.89)
- Factorized Diffusion Autoencoder for Unsupervised Disentangled Representation Learning (AAAI 2024 · similarity 0.86)
- Synergy Between Sufficient Changes and Sparse Mixing Procedure for Disentangled Representation Learning (ICLR 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.