Research map / Optimization / Implicit bias and generalization
Implicit Regularization for Group Sparsity
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 the diagonally grouped linear reparameterization with weight normalization | DGLNN on simulated group-sparse linear regression · Simulated data | recovery error | Helped |
| Stop training early | DGLNN on simulated group-sparse linear regression · Simulated data | structured-sparse recovery | Helped |
| Use group-structured sparsity rather than the standard-sparsity parameterization | Sparse linear regression with group size 4 · Simulated data, n=100, p=500 | convergence | Helped |
| Apply the proposed reparameterization in the size-one-group case | DGLNN · Simulated sparse linear regression, n=80, p=200 | recovery error | Helped |
Closest papers by meaning
- Implicit Sparse Regularization: The Impact of Depth and Early Stopping (NeurIPS 2021 · similarity 0.91)
- Implicit Regularization for Optimal Sparse Recovery (NeurIPS 2019 · similarity 0.89)
- Implicit Gradient Regularization (ICLR 2021 · similarity 0.89)
- Implicit Regularization for Tubal Tensor Factorizations via Gradient Descent (ICML 2025 · similarity 0.88)
- Implicit Regularization of Decentralized Gradient Descent for Sparse Regression (NeurIPS 2024 · similarity 0.88)
- Mask in the Mirror: Implicit Sparsification (ICLR 2025 · similarity 0.88)
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.