Research map / Optimization / Implicit bias and generalization
How Data Augmentation affects Optimization for Linear Regression
Boris Hanin, Yi Sun
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 zero noise variance (no augmentation) schedule in augmented GD/SGD (ablation) | augmented GD/SGD with Gaussian noise · synthetic linear regression, N=100, n=400 | orthogonal component ||W_perp|| and MSE | Hurt |
| Use exponentially decaying noise variance schedule (ablation) | augmented GD/SGD with Gaussian noise · synthetic linear regression | orthogonal component ||W_perp|| and MSE | Hurt |
| Use constant noise variance schedule (ablation) | augmented GD/SGD with Gaussian noise · synthetic linear regression | MSE | Hurt |
| Decay both noise strength and learning rate polynomially to zero at slow rates (ablation) | augmented GD/SGD with Gaussian noise · synthetic linear regression | MSE and ||W_perp|| | Improved |
Closest papers by meaning
- How Much Data Are Augmentations Worth? An Investigation into Scaling Laws, Invariance, and Implicit Regularization (ICLR 2023 · similarity 0.88)
- A Group-Theoretic Framework for Data Augmentation (NeurIPS 2020 · similarity 0.88)
- Data-Efficient Augmentation for Training Neural Networks (NeurIPS 2022 · similarity 0.87)
- Characterizing the Generalization Error of Random Feature Regression with Arbitrary Data-Augmentation (NeurIPS 2026 · similarity 0.85)
- Principled Data Augmentation for Learning to Solve Quadratic Programming Problems (NeurIPS 2025 · similarity 0.85)
- Risk Bounds of Accelerated SGD for Overparameterized Linear Regression (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.
