Research map / Learning theory and algorithms / Optimal transport
A Wasserstein Minimax Framework for Mixed Linear Regression
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 WMLR instead of EM | Centralized MLR · SNR 10, n=100,000 | NLL: 2.115 -> 2.059 | Helped |
| Use WMLR instead of GEM | Centralized MLR · SNR 10, n=100,000 | Relative ℓ2 error: 1.03 -> 5.31\times 10^{-3} | Helped |
| Use WMLR instead of EM | Centralized MLR · SNR 1, n=100,000 | NLL: 1.657 -> 1.656 | No clear effect |
| Use WMLR instead of GEM | Centralized MLR · SNR 1, n=100,000 | Relative ℓ2 error: 5.20\times 10^{-2} -> 7.78\times 10^{-2} | Hurt |
| Use WMLR instead of EM | Centralized MLR · SNR 10, n=10,000 | NLL: 2.715 -> 2.065 | Helped |
| Use WMLR instead of GEM | Centralized MLR · SNR 10, n=10,000 | Relative ℓ2 error: 9.98 \times 10^{-1} -> 2.08\times 10^{-2} | Helped |
| Use WMLR instead of GEM | Centralized MLR · SNR 1, n=10,000 | NLL: 1.657 -> 1.668 | Hurt |
| Use WMLR instead of GEM | Centralized MLR · SNR 1, n=10,000 | Relative ℓ2 error: 1.80 \times 10^{-1} -> 2.75\times 10^{-1} | Hurt |
| Use WMLR instead of GEM | Federated MLR · SNR 20 | Iterations required: 12,948 -> 74 | Helped |
| Use WMLR instead of GEM | Federated MLR · SNR 20 | Relative ℓ2 error: 1.93\times10^{-3} -> 2.49\times10^{-3} | Hurt |
| Use WMLR instead of GEM | Federated MLR · SNR 10 | Iterations required: 2,007 -> 98 | Helped |
| Use WMLR instead of GEM | Federated MLR · SNR 10 | Relative ℓ2 error: 3.92\times10^{-3} -> 4.93\times10^{-3} | Hurt |
Closest papers by meaning
- Convergence of Online Learning Algorithm for a Mixture of Multiple Linear Regressions (ICML 2024 · similarity 0.86)
- Unveiling the Cycloid Trajectory of EM Iterations in Mixed Linear Regression (ICML 2024 · similarity 0.84)
- MixMax: Distributional Robustness in Function Space via Optimal Data Mixtures (ICLR 2025 · similarity 0.84)
- Agnostic Learning of Mixed Linear Regressions with EM and AM Algorithms (ICML 2024 · similarity 0.84)
- SWAP: Sparse Entropic Wasserstein Regression for Robust Network Pruning (ICLR 2024 · similarity 0.84)
- SparsyFed: Sparse Adaptive Federated Learning (ICLR 2025 · similarity 0.84)
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.