Research map / Learning theory and algorithms / Sample complexity
Mixture Proportion Estimation Beyond Irreducibility
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 |
|---|---|---|---|
| Apply SuMPE subsampling to DPL | DPL · Gamma Ray | average absolute estimation error: 0.045+ -> 0.013+ | Helped |
| Apply SuMPE subsampling to EN | EN · Gamma Ray | average absolute estimation error: 0.034+ -> 0.027- | Helped |
| Apply SuMPE subsampling to KM | KM · Gamma Ray | average absolute estimation error: 0.163+ -> 0.042+ | Helped |
| Apply SuMPE subsampling to TIcE | TIcE · Gamma Ray | average absolute estimation error: 0.095+ -> 0.019+ | Helped |
| Apply SuMPE subsampling to DPL | DPL · Synthetic domain adaptation | average absolute estimation error: 0.060+ -> 0.028- | Helped |
| Apply SuMPE subsampling to EN | EN · Synthetic domain adaptation | average absolute estimation error: 0.045- -> 0.067- | Hurt |
| Apply SuMPE subsampling to KM | KM · Synthetic domain adaptation | average absolute estimation error: 0.063+ -> 0.022- | Helped |
| Apply SuMPE subsampling to TIcE | TIcE · Synthetic domain adaptation | average absolute estimation error: 0.128+ -> 0.041+ | Helped |
| Use regrouping instead of the original estimator | DPL, EN, KM, and TIcE · Gamma Ray | average absolute estimation error: 0.045+ -> 0.117-; 0.034+ -> 0.118-; 0.163+ -> 0.076+; 0.095+ -> 0.061+ | Mixed |
| Use regrouping instead of the original estimator | DPL, EN, KM, and TIcE · Synthetic domain adaptation | average absolute estimation error: 0.060+ -> 0.053+; 0.045- -> 0.061-; 0.063+ -> 0.059+; 0.128+ -> 0.094+ | Mixed |
| Regroup probability mass from F into H | ReMPE-2 under irreducibility · Mixture proportion estimation | asymptotic bias | Hurt |
Closest papers by meaning
- An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption (CVPR 2018 · similarity 0.91)
- Mixture Proportion Estimation via Kernel Embeddings of Distributions (ICML 2016 · similarity 0.90)
- Efficient Mixture Learning in Black-Box Variational Inference (ICML 2024 · similarity 0.82)
- Density Ratio Estimation with Doubly Strong Robustness (ICML 2024 · similarity 0.82)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.81)
- Subtractive Mixture Models via Squaring: Representation and Learning (ICLR 2024 · similarity 0.81)
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.