Research map / Learning theory and algorithms / Sample complexity
Gaussian Mixture Models with Rare Events
Xuetong Li, Jing Zhou, Hansheng Wang
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 |
|---|---|---|---|
| add 10% partially labeled data (MEM instead of EM) | EM for Gaussian mixture · Simulated GMM, alpha=10% | EM iterations N_iter: 176.60 -> 62.54 | Improved |
| add 10% partially labeled data (MEM instead of EM) | EM for Gaussian mixture · Simulated GMM, alpha=1% | EM iterations N_iter: 436.16 -> 75.00 | Improved |
| add 50% partially labeled data | EM for Gaussian mixture · Simulated GMM, alpha=0.1% | EM iterations N_iter: 733.78 -> 16.39 | Improved |
| add 10% partially labeled data | EM for Gaussian mixture · Simulated GMM, alpha=1% | spectral radius: 0.9839 -> 0.8852 | Improved |
| add 10% partially labeled data | EM for Gaussian mixture · Simulated GMM, alpha=1% | RMSE: 0.0663 -> 0.0306 | Improved |
| add 50% partially labeled data | EM for Gaussian mixture · Simulated GMM, alpha=0.1% | RMSE: 0.2668 -> 0.0593 | Improved |
| make the response rarer (alpha 50% -> 1%) without labels | EM for Gaussian mixture · EM | spectral radius: 0.9323 -> 0.9839 | Hurt |
| make the response rarer (alpha 50% -> 1%) without labels | EM for Gaussian mixture · EM | RMSE: 0.0092 -> 0.0663 | Hurt |
| make the response rarer (alpha 50% -> 0.1%) without labels | EM for Gaussian mixture · EM | EM iterations N_iter: 119.64 -> 733.78 | Hurt |
| add 25% labeled data | EM for Gaussian mixture · Swedish Traffic Signs | AUC: 0.5235 -> 0.9547 | Improved |
| use fully labeled instead of 25% labeled | EM for Gaussian mixture · Swedish Traffic Signs | AUC: 0.9547 -> 0.9711 | Improved |
| add 50% labeled data | EM for Gaussian mixture · Swedish Traffic Signs | false positives: 1,098.85 -> 5.45 | Improved |
Closest papers by meaning
- Toward Global Convergence of Gradient EM for Over-Paramterized Gaussian Mixture Models (NeurIPS 2024 · similarity 0.89)
- Local Maxima in the Likelihood of Gaussian Mixture Models: Structural Results and Algorithmic Consequences (NeurIPS 2016 · similarity 0.89)
- Convergence of Gradient EM on Multi-component Mixture of Gaussians (NeurIPS 2017 · similarity 0.89)
- Global Analysis of Expectation Maximization for Mixtures of Two Gaussians (NeurIPS 2016 · similarity 0.87)
- Big Learning Expectation Maximization (AAAI 2024 · similarity 0.87)
- Is $\sqrt{d}$ Separation Necessary for Gradient EM to Learn Gaussian Mixtures in High Dimensions? (NeurIPS 2026 · similarity 0.85)
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.
