Research map / Learning theory and algorithms / Sample complexity
An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption
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 MPEIA rather than ED for estimating the class prior | PNU(SL) on Magic · Magic | misclassification rate: 17.7 (1.8) -> 16.4 (1.0) | Helped |
| Use MPEIA rather than ED for estimating the class prior | PNU(SL) on SUSY · SUSY | estimated positive class prior: 0.3992 -> 0.2040 | Helped |
| Use MPEIA rather than KM for estimating the class prior | PNU(SL) on SUSY · SUSY | misclassification rate: 20.2 (0.3) -> 24.4 (2.4) | Hurt |
| Use MPEIA rather than ED for estimating the class prior | PNU(SL) on Waveform · Waveform | estimated positive class prior: 0.3407 -> 0.1892 | Helped |
| Use MPEIA rather than ED for estimating the class prior | PNU(SL) on Waveform · Waveform | misclassification rate: 15.2 (0.9) -> 9.2 (1.4) | Helped |
| Use MPEIA rather than KM for estimating the class prior | PNU(SL) on ijcnn1 · ijcnn1 | estimated positive class prior: 0.1096 -> 0.4509 | Hurt |
| Use MPEIA rather than ED for estimating the class prior | PNU(SL) on ijcnn1 · ijcnn1 | misclassification rate: 36.9 (5.8) -> 36.4 (5.4) | No clear effect |
| Use MPEIA rather than ED for estimating the class prior | PNU(SL) on Spambase · Spambase | estimated positive class prior: 0.3178 -> 0.2909 | Helped |
| Use MPEIA rather than PE for estimating the class prior | PNU(SL) on Spambase · Spambase | misclassification rate: 13.3 (2.6) -> 14.8 (0.9) | Hurt |
| Use MPEIA rather than KM for estimating the class prior | PNU(SL) on a9a · a9a | estimated positive class prior: 0.2062 -> 0.1990 | Helped |
| Use MPEIA rather than KM for estimating the class prior | PNU(SL) on a9a · a9a | misclassification rate: 18.1 (1.1) -> 18.1 (0.9) | No clear effect |
| Use MPEIA rather than KM for estimating the class prior | PNU(SL) on w8a · w8a | estimated positive class prior: 0.3109 -> 0.2833 | Helped |
Closest papers by meaning
- Mixture Proportion Estimation via Kernel Embeddings of Distributions (ICML 2016 · similarity 0.92)
- Mixture Proportion Estimation Beyond Irreducibility (ICML 2023 · similarity 0.91)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.83)
- Efficient Mixture Learning in Black-Box Variational Inference (ICML 2024 · similarity 0.82)
- Score-of-Mixture Training: One-Step Generative Model Training Made Simple via Score Estimation of Mixture Distributions (ICML 2025 · similarity 0.82)
- Be More Diverse than the Most Diverse: Optimal Mixtures of Generative Models via Mixture-UCB Bandit Algorithms (ICLR 2025 · similarity 0.82)
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.