Research map / Optimization / Optimizers
Globally aware optimization with resurgence
Wei Bu
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 |
|---|---|---|---|
| apply Borel-Pade resummation instead of truncated asymptotic series (ablation) | partition function approximation · Quartic partition function Z(g), g=0.01 | Z(g) value: 0.4524 -> 0.5605 (exact 0.5606) | Improved |
| apply Borel-Pade resummation instead of truncated asymptotic series (ablation) | partition function approximation · Quartic partition function Z(g), g=0.005 | Z(g) value: 0.2883 -> 0.3959 (exact 0.3960) | Improved |
| apply Borel-Pade resummation instead of truncated asymptotic series (ablation) | partition function approximation · Quartic partition function Z(g), g=0.1 | Z(g) value: 1.7023 -> 1.7721 (exact 1.7724) | Improved |
| add non-perturbative trans-series corrections to Borel-Pade (ablation) | partition function approximation · Quartic partition function Z(g), g=0.1 | Z(g) value: 1.7721 -> 1.7724 | Improved |
| add non-perturbative trans-series corrections to Borel-Pade (ablation) | partition function approximation · Quartic partition function Z(g), g=0.005 | Z(g) value: 0.3959 -> 0.3960 | Improved |
| wrap SGD, Adam, AdamW or Muon with SURGE learning-rate scaling | base optimizers · MNIST MLP and Shakespeare small transformer | training loss | Improved |
| wrap base optimizers with SURGE learning-rate scaling | base optimizers · MNIST MLP and Shakespeare small transformer | training stability | Hurt |
| wrap an optimizer that generalizes poorly with SURGE | base optimizers · MNIST MLP and Shakespeare small transformer | generalization | Hurt |
Closest papers by meaning
- Adaptive Partitioning Schemes for Optimistic Optimization (ICML 2025 · similarity 0.85)
- Sharper Convergence Rates for Nonconvex Optimisation via Reduction Mappings (NeurIPS 2025 · similarity 0.85)
- Optimizing Posterior Samples for Bayesian Optimization via Rootfinding (ICLR 2025 · similarity 0.85)
- ROOT: Rethinking Offline Optimization as Distributional Translation via Probabilistic Bridge (NeurIPS 2025 · similarity 0.85)
- LoRA Training Provably Converges to a Low-Rank Global Minimum Or It Fails Loudly (But it Probably Won't Fail) (ICML 2025 · similarity 0.85)
- Gradient Descent as Loss Landscape Navigation: a Normative Framework for Deriving Learning Rules (NeurIPS 2025 · 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.
