Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Distributed Learning of Finite Gaussian Mixtures
Qiong Zhang, Jiahua Chen
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 GMR reduction aggregation instead of the global pMLE | Finite Gaussian mixture learning · MAGIC04 | log-likelihood per observation: -24.15 -> -24.30(0.07) | Hurt |
| Use GMR reduction aggregation instead of the global pMLE | Finite Gaussian mixture learning · MiniBooNE | log-likelihood per observation: -19.46 -> -22.00(0.53) | Hurt |
| Use GMR reduction aggregation instead of the global pMLE | Finite Gaussian mixture learning · KDD | log-likelihood per observation: -221.80 -> -223.25(0.42) | Hurt |
| Use GMR reduction aggregation instead of the global pMLE | Finite Gaussian mixture learning · MSYP | log-likelihood per observation: -166.56 -> -167.05(0.04) | Hurt |
| Use GMR split-and-conquer learning instead of global learning | Finite Gaussian mixture learning · MAGIC04 | computational time (seconds): 19.3 -> 7.0(3.2) | Improved |
| Use GMR split-and-conquer learning instead of global learning | Finite Gaussian mixture learning · MiniBooNE | computational time (seconds): 346.9 -> 313.1(162.6) | Improved |
| Use GMR split-and-conquer learning instead of global learning | Finite Gaussian mixture learning · KDD | computational time (seconds): 1033.9 -> 544.4(309.5) | Improved |
| Use GMR split-and-conquer learning instead of global learning | Finite Gaussian mixture learning · MSYP | computational time (seconds): 67048.8 -> 2611.6(474.0) | Improved |
| Increase the number of local machines | GMR · Simulated GMMs | W1, ARI, and log-likelihood | No clear effect |
| Increase the number of local machines | Median aggregation · Simulated GMMs | W1 and ARI | Hurt |
| Increase the number of local machines | Coreset aggregation · Simulated GMMs | W1 and ARI | Hurt |
| Increase total sample size | Coreset aggregation with fixed coreset size · Simulated GMMs | W1 and ARI | Hurt |
Closest papers by meaning
- Sample-Efficient Private Learning of Mixtures of Gaussians (NeurIPS 2024 · similarity 0.84)
- Convergence of Gradient EM on Multi-component Mixture of Gaussians (NeurIPS 2017 · similarity 0.84)
- Differentially Private Algorithms for Learning Mixtures of Separated Gaussians (NeurIPS 2019 · similarity 0.84)
- Toward Global Convergence of Gradient EM for Over-Paramterized Gaussian Mixture Models (NeurIPS 2024 · similarity 0.83)
- Learning Mixtures of Gaussian Processes through Random Projection (ICML 2024 · similarity 0.83)
- On Learning Parallel Pancakes with Mostly Uniform Weights (ICML 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.
