Research map / Learning theory and algorithms / Prediction and decision losses
Bilevel learning of the Group Lasso structure
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 |
|---|---|---|---|
| learn group structure with bilevel method (BiGL) instead of using identity groups (Lasso) | Group Lasso · Gene ontology classes from regulatory motifs | prediction error | Helped |
| learned groups (BiGL) vs oracle Group Lasso with true groups | Group Lasso · Synthetic, N=50, P=100, T=500 | estimation error | No clear effect |
| increase number of tasks T (ablation) | BiGL · Synthetic | estimation error | Helped |
| increase number of inner iterations Q (ablation) | BiGL · Synthetic | validation error | Helped |
| set maximum number of groups L=20 instead of true 10 (ablation) | BiGL · Synthetic | group recovery: 9 of 10 extra groups empty | No clear effect |
| threshold groups to dominant group (BiGLThr) vs soft BiGL (ablation) | BiGL · Synthetic | validation error | No clear effect |
Closest papers by meaning
- Smooth Bilevel Programming for Sparse Regularization (NeurIPS 2021 · similarity 0.87)
- Chunk Dynamic Updating for Group Lasso with ODEs (AAAI 2022 · similarity 0.87)
- GAP Safe Screening Rules for Sparse-Group Lasso (NeurIPS 2016 · similarity 0.87)
- Dual Feature Reduction for the Sparse-group Lasso and its Adaptive Variant (ICML 2025 · similarity 0.86)
- qNBO: quasi-Newton Meets Bilevel Optimization (ICLR 2025 · similarity 0.84)
- Bilevel Optimization under Unbounded Smoothness: A New Algorithm and Convergence Analysis (ICLR 2024 · 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.