Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Influence-Augmented Online Planning for Complex Environments
Jinke He, Miguel Suau de Castro, Frans Oliehoek
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 the influence predictor with a GRU instead of a uniform random predictor in the IALS (ablation) | IALS with POMCP · Grab A Chair, 5 to 129 agents | average return | Improved |
| Replace global simulator with IALS using a GRU influence predictor, same simulations per step | POMCP · Grab A Chair | average return | No clear effect |
| Replace global simulator with IALS, many agents | POMCP · Grab A Chair, many agents | simulation time per step | Improved |
| Replace global simulator with IALS, few agents | POMCP · Grab A Chair, few agents | simulation time per step | Hurt |
| Lower probability p of winning contested chair (stronger influence), IALS vs global | IALS with GRU predictor · 5-agent Grab A Chair variant | average return gap to global simulator | Hurt |
| Replace global simulator with IALS under constrained planning time | POMCP · Grid Traffic Control | discounted return | Improved |
| Replace global simulator with IALS under more planning time | POMCP · Grid Traffic Control | discounted return | No clear effect |
| Replace global simulator with IALS | POMCP · Grid Traffic Control | number of simulations per planning step | Improved |
| Replace global simulator with IALS | POMCP · Grid Traffic Control | steps before particle depletion | Improved |
Closest papers by meaning
- Factored Online Planning in Many-Agent POMDPs (AAAI 2024 · similarity 0.85)
- Graph Sparse Sampling: Breaking the Curse of the Horizon in Continuous MDP Planning (NeurIPS 2026 · similarity 0.83)
- Planning, Fast and Slow: Online Reinforcement Learning with Action-Free Offline Data via Multiscale Planners (ICML 2024 · similarity 0.83)
- Wide-Horizon Thinking and Simulation-Based Evaluation for Real-World LLM Planning with Multifaceted Constraints (NeurIPS 2025 · similarity 0.83)
- Latent Plan Transformer for Trajectory Abstraction: Planning as Latent Space Inference (NeurIPS 2024 · similarity 0.83)
- Resisting Stochastic Risks in Diffusion Planners with the Trajectory Aggregation Tree (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.
