Research map / Learning theory and algorithms / Combinatorial and search optimization
Improving Search with Supervised Learning in Trick-Based Card Games
Christopher Solinas, Douglas Rebstock, Michael Buro
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 human-trained cardplay-history inference in addition to bidding and declaration inference | PIMC Skat player · Suit games | average tournament score per game: 4.9; 5.0; 5.4 | Improved |
| Use human-trained cardplay-history inference in addition to bidding and declaration inference | PIMC Skat player · Grand games | average tournament score per game: 2.6; 2.9; 3.1 | Improved |
| Use human-trained cardplay-history inference in addition to bidding and declaration inference | PIMC Skat player · Null games | average tournament score per game: 6.3; 6.3; 5.0 | Improved |
| Use neural card-location inference from bidding, declaration, and cardplay history instead of uniform sampling | PIMC Skat player · Suit games | average tournament score per game: 9.4; 7.6; 8.3 | Improved |
| Use neural card-location inference from bidding, declaration, and cardplay history instead of uniform sampling | PIMC Skat player · Null games | average tournament score per game: 8.9; 9.9; 10.8 | Improved |
| Use neural card-location inference from bidding, declaration, and cardplay history instead of uniform sampling | PIMC Skat player · Grand games | average tournament score per game: 3.7; 3.7; 4.0 | Improved |
| Use cardplay-history inference instead of BDI's bidding-and-declaration-only inference | PIMC Skat player · Suit games | soloist win percentage | Improved |
| Use cardplay-history inference instead of BDI's bidding-and-declaration-only inference | PIMC Skat player · Null games | soloist win percentage | Improved |
| Use cardplay-history inference instead of BDI's bidding-and-declaration-only inference | PIMC Skat player · Grand games | average tournament score per game against KI: 0.9; 1.3; 2.1 | No clear effect |
| Use Kermit's count-based inference instead of BDI individual-card predictions without cardplay history | PIMC Skat player · Suit games | average tournament score per game: 2.1; 0.5; 0.4 | Improved |
| Use Kermit's count-based inference instead of BDI individual-card predictions without cardplay history | PIMC Skat player · Grand games | average tournament score per game: 0.3; 1.4; 0.5 | Mixed |
| Use Kermit's count-based inference instead of BDI individual-card predictions without cardplay history | PIMC Skat player · Null games | average tournament score per game: 1.1; 1.3; 1.3 | Improved |
Closest papers by meaning
- Implicit Search via Discrete Diffusion: A Study on Chess (ICLR 2025 · similarity 0.85)
- Can Large Language Models Master Complex Card Games? (NeurIPS 2025 · similarity 0.85)
- Mastering Board Games by External and Internal Planning with Language Models (ICML 2025 · similarity 0.84)
- Amortized Planning with Large-Scale Transformers: A Case Study on Chess (NeurIPS 2024 · similarity 0.84)
- Learning Not to Regret (AAAI 2024 · similarity 0.84)
- Outbidding and Outbluffing Elite Humans: Mastering Liar’s Poker via Self-Play and Reinforcement Learning (NeurIPS 2026 · similarity 0.84)
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.
