Research map / Optimization / Optimizers
Learning to Optimize in Swarms
Yue Cao, Tianlong Chen, Zhangyang Wang, Yang Shen
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 |
|---|---|---|---|
| Run k independently initialized DM_LSTMs and select the best solution (ablation) | DM_LSTM · Rastrigin 10D | minimum objective: 55.4±13.5 -> 48.4±10.5 | No clear effect |
| Run k independently initialized DM_LSTMs and select the best solution (ablation) | DM_LSTM · Rastrigin 20D | minimum objective: 140.4±10.2 -> 137.4±12.7 | No clear effect |
| Add point-based features and intra-particle attention to independent particles (ablation) | B1 · Rastrigin 10D | minimum objective: 48.4±10.5 -> 40.1±9.4 | Improved |
| Add point-based features and intra-particle attention to independent particles (ablation) | B1 · Rastrigin 20D | minimum objective: 137.4±12.7 -> 108.4±13.4 | Improved |
| Add population-based features and inter-particle attention to create a swarm (ablation) | B2 · Rastrigin 10D | minimum objective: 40.1±9.4 -> 20.4±6.6 | Improved |
| Add population-based features and inter-particle attention to create a swarm (ablation) | B2 · Rastrigin 20D | minimum objective: 108.4±13.4 -> 48.5±7.1 | Improved |
| Add the posterior entropy term to the meta-loss (ablation) | B3 · Rastrigin 10D | minimum objective: 20.4±6.6 -> 12.3±5.4 | Improved |
| Add the posterior entropy term to the meta-loss (ablation) | B3 · Rastrigin 20D | minimum objective: 48.5±7.1 -> 43.0 ±9.2 | Improved |
Closest papers by meaning
- $\texttt{MetaDistill}$: Unlocking the Performance Ceiling for Pretrained Optimizers (ICML 2026 · similarity 0.88)
- MetaBox-v2: A Unified Benchmark Platform for Meta-Black-Box Optimization (NeurIPS 2025 · similarity 0.87)
- MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters (ICML 2025 · similarity 0.87)
- SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization (NeurIPS 2026 · similarity 0.86)
- Memory-Reduced Meta-Learning with Guaranteed Convergence (AAAI 2025 · similarity 0.86)
- A Generalization Result for Convergence in Learning-to-Optimize (ICML 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.
