Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Deep Disentangled Embeddings With the F-Statistic Loss
Karl Ridgeway, Michael Mozer
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 |
|---|---|---|---|
| Use the F-statistic loss instead of competing embedding losses | deep embeddings · CUHK03 | recall@1 accuracy | Improved |
| Use the F-statistic loss instead of competing embedding losses | deep embeddings · Market-1501 | recall@1 accuracy | No clear effect |
| Use the F-statistic loss instead of competing embedding losses | deep embeddings · CUB-200-2011 | recall@1 accuracy | Hurt |
| Train with the F-statistic loss rather than triplet or histogram loss | Sprites representations · Sprites | modularity | Improved |
| Train with the F-statistic loss rather than triplet or histogram loss | Small NORB representations · Small NORB | modularity | Improved |
| Train with the F-statistic loss rather than triplet or histogram loss | Small NORB representations · Small NORB | explicitness | Improved |
| Train with the F-statistic loss rather than triplet or histogram loss | Sprites representations · Sprites | explicitness | No clear effect |
| Increase β (ablation) | β-VAE · Sprites | modularity | Improved |
| Increase β (ablation) | β-VAE · Sprites | explicitness | Hurt |
| Replace the F-statistic with the Welch correction for a t statistic (ablation) | F-statistic loss · classification datasets | model performance | No clear effect |
| Use the unnamed-factor oracle rather than unsupervised β-VAE training | Sprites representations · Sprites | modularity | Improved |
| Use the class-aware oracle rather than unsupervised β-VAE training | Sprites representations · Sprites | explicitness | Improved |
Closest papers by meaning
- A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation (JMLR 2020 · similarity 0.89)
- Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations (ICML 2019 · similarity 0.89)
- Graph-based Unsupervised Disentangled Representation Learning via Multimodal Large Language Models (NeurIPS 2024 · similarity 0.88)
- Disentangled Representation Learning via Modular Compositional Bias (NeurIPS 2025 · similarity 0.88)
- Tripod: Three Complementary Inductive Biases for Disentangled Representation Learning (ICML 2024 · similarity 0.87)
- Enriching Disentanglement: From Logical Definitions to Quantitative Metrics (NeurIPS 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.
