Research map / Optimization / Implicit bias and generalization
Implicit Bias and Loss of Plasticity in Matrix Completion: Depth Promotes Low-Rankness
Baekrok Shin, Chulhee Yun
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 |
|---|---|---|---|
| increase depth of matrix factorization from 2 to 3 or more layers (ablation) | deep matrix factorization · 2x2 diagonal-observation completion (M_D), small init | effective rank of learned solution: rank-2 (L=2) -> rank-1 (L>=3) | Improved |
| increase depth for connected observation patterns (ablation) | deep matrix factorization · 2x2 connected observation (M_C) | low-rank bias | Improved |
| pre-train depth-2 network on few observations then resume with more data | depth-2 matrix completion · loss of plasticity | convergence to low rank | Hurt |
| use deeper network under pre-train then augment protocol (ablation) | deep matrix factorization · loss of plasticity | convergence to low rank | Improved |
Closest papers by meaning
- Connectivity Shapes Implicit Regularization in Matrix Factorization Models for Matrix Completion (NeurIPS 2024 · similarity 0.88)
- Deep Linear Network Training Dynamics from Random Initialization: Data, Width, Depth, and Hyperparameter Transfer (ICML 2025 · similarity 0.88)
- Feature Learning Dynamics in Infinite-Depth Neural Networks (NeurIPS 2026 · similarity 0.87)
- Mind the Gap: a Spectral Analysis of Rank Collapse and Signal Propagation in Attention Layers (ICML 2025 · similarity 0.86)
- Flatness is Necessary, Neural Collapse is Not: Rethinking Generalization via Grokking (NeurIPS 2025 · similarity 0.86)
- Analyzing Neural Scaling Laws in Two-Layer Networks with Power-Law Data Spectra (ICLR 2025 · similarity 0.86)
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.
