Research map / Learning theory and algorithms / Prediction and decision losses
Learning to Learn with Smooth Regularization
Yuanhao Xiong, Cho-Jui Hsieh
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 |
|---|---|---|---|
| add smoothness-inducing regularizer | SimpleOptimizer · CIFAR-10 3-layer CNN | test accuracy (%): 69.02 -> 72.50 | Improved |
| use PGD smoothness regularizer instead of l2 regularization (ablation) | SimpleOptimizer · CIFAR-10 3-layer CNN | test accuracy (%): 69.69 -> 72.50 | Improved |
| use PGD smoothness regularizer instead of spectral normalization (ablation) | SimpleOptimizer · CIFAR-10 3-layer CNN | test accuracy (%): 69.35 -> 72.50 | Improved |
| add orthogonal regularization (ablation) | SimpleOptimizer · CIFAR-10 3-layer CNN | test accuracy (%): 69.02 -> 10.00 | Hurt |
| add l2 regularization (ablation) | SimpleOptimizer · CIFAR-10 3-layer CNN | test accuracy (%): 69.02 -> 69.69 | No clear effect |
| add smoothness regularizer | Meta-LSTM · miniImageNet 5-way 1-shot | accuracy (%): 38.20 -> 40.42 | Improved |
| add smoothness regularizer | Meta-LSTM · miniImageNet 5-way 5-shot | accuracy (%): 56.56 -> 58.90 | Improved |
| add smoothness regularizer | Meta-LSTM · tieredImageNet 5-way 1-shot | accuracy (%): 36.43 -> 36.74 | No clear effect |
| add smoothness regularizer | Meta-LSTM · tieredImageNet 5-way 5-shot | accuracy (%): 53.45 -> 55.14 | Improved |
| add smoothness regularizer | SIB WRN-28-10 · miniImageNet 5-way 5-shot | accuracy (%): 78.9 -> 80.8 | Improved |
| add smoothness regularizer | SIB WRN-28-10 · miniImageNet 5-way 1-shot | accuracy (%): 69.6 -> 70.0 | No clear effect |
| add smoothness regularizer | SIB WRN-28-10 · CIFAR-FS 5-way 5-shot | accuracy (%): 85.3 -> 86.1 | Improved |
Closest papers by meaning
- Characterizing the Edge of Stability in Variational Training Without Priors (NeurIPS 2026 · similarity 0.87)
- Learning Continually by Spectral Regularization (ICLR 2025 · similarity 0.85)
- When Will Gradient Regularization Be Harmful? (ICML 2024 · similarity 0.85)
- On the Parameterization of Second-Order Optimization Effective towards the Infinite Width (ICLR 2024 · similarity 0.85)
- Variational Learning Finds Flatter Solutions at the Edge of Stability (NeurIPS 2025 · similarity 0.85)
- Investigation into Training Dynamics of Learned Optimizers (Student Abstract) (AAAI 2024 · similarity 0.85)
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.
