Research map / Learning theory and algorithms / Sample complexity
Imbalanced Mixed Linear Regression
Pini Zilber, Boaz Nadler
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 sequential robust-regression recovery instead of simultaneous component estimation | Mix-IRLS · Synthetic imbalanced mixtures | failure probability (F_latent > 2σ) | Improved |
| Use sequential robust-regression recovery | Mix-IRLS · Synthetic mixtures with outliers | maximum tolerated corrupted fraction: f=0.09 | Improved |
| Use sequential robust-regression recovery | Mix-IRLS · Balanced synthetic mixtures with outliers | maximum tolerated corrupted fraction: f=0.17 | Improved |
| Allow sequential component discovery rather than requiring a fixed correct component count | Mix-IRLS · Synthetic mixtures with overestimated K | recovery success | Improved |
| Tune Mix-IRLS parameters (ablation) | Mix-IRLS · Synthetic simulations | recovery performance | No clear effect |
| Tune Mix-IRLS parameters (ablation) | Mix-IRLS · Medical insurance, wine quality, WHO life expectancy, and fish market | F_real | Improved |
| Tune Mix-IRLS parameters (ablation) | Mix-IRLS · Wine quality (K=7) | F_real | Improved |
| Add the refinement phase after sequential recovery (ablation) | Mix-IRLS · Real-world datasets | F_real | Improved |
| Add the refinement phase after sequential recovery (ablation) | Mix-IRLS · Synthetic mixtures | F_latent | No clear effect |
| Provide the true mixture proportions as prior knowledge (ablation) | EM · Synthetic simulations | recovery performance | No clear effect |
Closest papers by meaning
- Agnostic Learning of Mixed Linear Regressions with EM and AM Algorithms (ICML 2024 · similarity 0.84)
- MixMin: Finding Data Mixtures via Convex Minimization (ICML 2025 · similarity 0.83)
- Linear Regression using Heterogeneous Data Batches (NeurIPS 2024 · similarity 0.83)
- Unveiling the Cycloid Trajectory of EM Iterations in Mixed Linear Regression (ICML 2024 · similarity 0.83)
- PRIME: Deep Imbalanced Regression with Proxies (ICML 2025 · similarity 0.83)
- Balanced Data, Imbalanced Spectra: Unveiling Class Disparities with Spectral Imbalance (ICML 2024 · similarity 0.83)
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.
