Research map / Learning theory and algorithms / Combinatorial and search optimization
Cost Effective Active Search
Shali Jiang, Roman Garnett, Benjamin Moseley
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 |
|---|---|---|---|
| ENCES-20 instead of GREEDY | ENCES · Drug discovery | average cost: 430.2 -> 152.2 | Improved |
| ENCES-20 instead of ENS-30 | ENCES · Drug discovery | average cost: 171.3 -> 152.2 | Improved |
| ENCES-0.2 instead of ENS-0.7 | ENCES · Drug discovery | average cost: 172.0 -> 170.7 | No clear effect |
| TWO-STEP instead of GREEDY | TWO-STEP · Drug discovery | average cost: 430.2 -> 198.4 | Improved |
| TWO-STEP instead of GREEDY | TWO-STEP · Materials discovery | average cost: 972.9 -> 976.6 | No clear effect |
| ENCES-0.5 instead of GREEDY | ENCES · Materials discovery | average cost: 972.9 -> 897.0 | Improved |
| ENCES-50 instead of ENS-70 | ENCES · Materials discovery | average cost: 910.8 -> 897.1 | Improved |
| ENCES-0.5 at T=50 vs GREEDY | ENCES · Materials discovery T=50 | average cost: 84.5 -> 87.1 | Hurt |
| set r as constant 20 vs 0.2 fraction of remaining target at T=200 | ENCES · Drug discovery | average cost: 255.1 -> 298.9 | Mixed |
| ACCU' approximation instead of EXACT expectation | ACCU' · NPB expectation | time (s): 97.4841 -> 0.0029 | Improved |
| RECIP approximation instead of ACCU' | RECIP · NPB expectation | RMSE: 0.0438 -> 7.8276 | Hurt |
| ACCU instead of ACCU' | ACCU · NPB expectation | RMSE: 0.0438 -> 0.5723 | Hurt |
Closest papers by meaning
- Efficient Nonmyopic Active Search (ICML 2017 · similarity 0.91)
- Efficient nonmyopic batch active search (NeurIPS 2018 · similarity 0.90)
- Training Greedy Policy for Proposal Batch Selection in Expensive Multi-Objective Combinatorial Optimization (ICML 2024 · similarity 0.84)
- Uncertainty Herding: One Active Learning Method for All Label Budgets (ICLR 2025 · similarity 0.84)
- ALINE: Joint Amortization for Bayesian Inference and Active Data Acquisition (NeurIPS 2025 · similarity 0.83)
- Random Exploration in Bayesian Optimization: Order-Optimal Regret and Computational Efficiency (ICML 2024 · similarity 0.83)
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.
