Scholarlink

Physics-informed neural networks (PINNs) are a promising machine learning approach for solving the difficult-to-solve partial differential equations that govern differential games. However, the application of PINNs to multiplayer differential games remains an open research gap due to their nascency. In this work, we provide a brief overview of how PINNs can be utilised to solve adversarial differential games involving multiple independent players.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

(Poster) Physics-Informed Value Approximation for Pursuit-Evasion Games

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Takuma Adams,
  • Andrew C. Cullen,
  • Tansu Alpcan

摘要

Physics-informed neural networks (PINNs) are a promising machine learning approach for solving the difficult-to-solve partial differential equations that govern differential games. However, the application of PINNs to multiplayer differential games remains an open research gap due to their nascency. In this work, we provide a brief overview of how PINNs can be utilised to solve adversarial differential games involving multiple independent players.