Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Bayesian Learning of Other Agents' Finite Controllers for Interactive POMDPs
Alessandro Panella, Piotr Gmytrasiewicz
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 |
|---|---|---|---|
| learn PDFC models of agent j instead of using a uniform-random model of j's actions | Agent i in I-POMDP · Tiger, 3x4 Maze, 5x5 AUAV | agent i mean total reward | Improved |
| learn PDFC models of agent j instead of predicting j's actions by long-term frequency | Agent i in I-POMDP · Tiger, 3x4 Maze, 5x5 AUAV | agent i mean total reward | Improved |
| increase length of training observation sequence | Learned PDFC models · Tiger, 3x4 Maze, 5x5 AUAV | agent i mean total reward | Improved |
| use only 256 observations for learning on Maze | Learned PDFC models · 3x4 Maze | fraction of reward gap closed: about 90% | Improved |
| increase Tlearn with perfect observability | PDFC learning · Tiger | wKL to true controller: Tlearn >= 64 | Improved |
| increase Tlearn with perfect observability | PDFC learning · 3x4 Maze | number of nodes of learned PDFC: below true size 42 | Mixed |
| increase Tlearn with perfect observability | PDFC learning · 5x5 AUAV | number of nodes of learned PDFC: approaches 36 | Improved |
| imperfect observability of j instead of perfect observability | PDFC learning · all three problems | size of learned PDFC | Hurt |
Closest papers by meaning
- Sequential Monte Carlo for Policy Optimization in Continuous POMDPs (NeurIPS 2025 · similarity 0.87)
- Learning POMDP World Models from Observations with Language-Model Priors (NeurIPS 2026 · similarity 0.87)
- Periodic agent-state based Q-learning for POMDPs (NeurIPS 2024 · similarity 0.87)
- Sample-Efficient Learning of POMDPs with Multiple Observations In Hindsight (ICLR 2024 · similarity 0.87)
- Factored Online Planning in Many-Agent POMDPs (AAAI 2024 · similarity 0.86)
- Multi-Environment POMDPs with Finite-Horizon Objectives (NeurIPS 2026 · 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.
