Research map / Learning theory and algorithms / Submodular and game algorithms
Adaptive Maximization of Pointwise Submodular Functions With Budget Constraint
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 combined policy (best-of cost-average and cost-insensitive greedy, each with half budget) instead of either greedy alone | budgeted pool-based active learning · Table 1 across 3 datasets and 4 cost scenarios | AUC | Mixed |
| Cost-average greedy (ALC) vs cost-insensitive greedy (LC) under random/margin costs | budgeted pool-based active learning · Table 1 | AUC | Mixed |
| Pointwise submodularity assumption: greedy policies not near-optimal with non-uniform modular cost (theory) | cost-average greedy and cost-insensitive greedy · worst-case adaptive optimization, modular cost | worst-case utility ratio | Hurt |
| Take best of the two greedy policies (pi = argmax of worst-case utility) instead of either alone | worst-case adaptive optimization with modular cost · theoretical guarantee | approximation ratio: guarantee 1/2(1-1/e) vs. none for individual policies | Helped |
| Cost-insensitive greedy with full budget compared against optimal half-budget policy | cost-insensitive greedy policy · theory, modular cost | approximation ratio | Hurt |
Closest papers by meaning
- Beyond Adaptive Submodularity: Approximation Guarantees of Greedy Policy with Adaptive Submodularity Ratio (ICML 2019 · similarity 0.89)
- Randomized Algorithms for Submodular Function Maximization with a $k$-System Constraint (ICML 2021 · similarity 0.89)
- Weakly Submodular Maximization Beyond Cardinality Constraints: Does Randomization Help Greedy? (ICML 2018 · similarity 0.89)
- Consistent Submodular Maximization (ICML 2024 · similarity 0.89)
- Online Submodular Maximization via Online Convex Optimization (AAAI 2024 · similarity 0.88)
- Nearly Minimax Optimal Submodular Maximization with Bandit Feedback (NeurIPS 2024 · similarity 0.87)
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.