Research map / Learning theory and algorithms / Combinatorial and search optimization
GAGA: Deciphering Age-path of Generalized Self-paced Regularizer
Xingyu Qu, Diyang Li, Xiaohan Zhao, Bin Gu
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 GAGA with the linear SP-regularizer instead of the strongest listed baseline | SVM · mfeat-pixel (30% noise) | accuracy: 0.978 -> 0.986 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | SVM · mfeat-pixel (20% noise) | accuracy: 0.960 -> 0.983 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | SVM · hiva agnostic (30% noise) | accuracy: 0.946 -> 0.960 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | SVM · pendigits (30% noise) | accuracy: 0.962 -> 0.971 | Improved |
| Use GAGA with the linear SP-regularizer instead of ACS | SVM · pendigits (20% noise) | accuracy: 0.942 -> 0.944 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | Lasso · elevator (30% noise) | generalization error: 0.144 -> 0.143 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | Lasso · ailerons (30% noise) | generalization error: 0.491 -> 0.489 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | Lasso · music (30% noise) | generalization error: 0.215 -> 0.206 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | Lasso · delta elevators (30% noise) | generalization error: 0.679 -> 0.634 | Improved |
| Use GAGA with the linear SP-regularizer instead of the strongest listed baseline | Lasso · houses (30% noise) | generalization error: 0.205 -> 0.201 | Improved |
| Use GAGA with the mixture SP-regularizer instead of the strongest listed baseline | SVM · mfeat-pixel (30% noise) | accuracy: 0.968 -> 0.973 | Improved |
| Use GAGA with the mixture SP-regularizer instead of the strongest listed baseline | SVM · mfeat-pixel (20% noise) | accuracy: 0.970 -> 0.977 | Improved |
Closest papers by meaning
- PACE: Marrying generalization in PArameter-efficient fine-tuning with Consistency rEgularization (NeurIPS 2024 · similarity 0.80)
- Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors (ICLR 2024 · similarity 0.80)
- Symbolic regression via MDLformer-guided search: from minimizing prediction error to minimizing description length (ICLR 2025 · similarity 0.80)
- Compress to Impress: Efficient LLM Adaptation Using a Single Gradient Step on 100 Samples (NeurIPS 2025 · similarity 0.80)
- Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regularization in Training (NeurIPS 2025 · similarity 0.80)
- Improving Training Efficiency of Diffusion Models via Multi-Stage Framework and Tailored Multi-Decoder Architecture (CVPR 2024 · similarity 0.79)
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.
