A Player Selection Network for Scalable Game-Theoretic Prediction and Planning

TMLR Paper8216 Authors

Abstract: While game-theoretic planning frameworks are effective at modeling multi-agent interactions, they require solving large optimization problems where the number of variables increases with the number of agents, resulting in long computation times that limit their use in large-scale, real-time systems. To address this issue, we propose i) PSN Game—a learning-based, game-theoretic prediction and planning framework that reduces game size by learning a Player Selection Network (PSN); and ii) a Goal Inference Network (GIN) that makes it possible to use the PSN in incomplete-information games where other agents’ intentions are unknown to the ego agent. A PSN outputs a player selection mask that distinguishes influential players from less relevant ones, enabling the ego player to solve a smaller, masked game involving only selected players. By reducing the number of players included in the game, PSN shrinks the corresponding optimization problems, leading to faster solve times. The PSN Game framework is more flexible than existing player selection methods as it i) relies solely on observations of players’ past trajectories, without requiring full state, action, or other game-specific information; and ii) requires no online parameter tuning. Experiments in both simulated scenarios and real-world pedestrian trajectory datasets show that PSN is competitive with, and often improves upon, the evaluated explicit game-theoretic selection baselines in i) prediction accuracy and ii) planning safety. Across scenarios, PSN typically selects substantially fewer players than are present in the full game, thereby reducing game size and planning complexity. PSN also generalizes to settings in which agents’ objectives are unknown, via the GIN, without test-time fine-tuning. By selecting only the most relevant players for decision-making, PSN Game provides a practical mechanism for reducing planning complexity that can be integrated into existing multi-agent planning frameworks.

Submission Type: Regular submission (no more than 12 pages of main content)

Changes Since Last Submission:

In general, we add the following content requested by the reviewers.
1. In the main text, we add the runtime analysis for PSN and GIN, and the computational time statistics for all results.
2. In the appendix, we add the mask distribution output by PSN, which supports the sensitivity analysis for the choice of threshold.
3. We add remark 2 to better explain the role of the pre-filter for large-scale scenarios.
4. Some minor presentation issues raised by the reviewers.

We highlight the changes in the revised manuscript with blue color. Please feel free to refer to them.

Assigned Action Editor: ~Chen_Sun1

Submission Number: 8216