Scholarlink

Reinforcement Learning (RL) [10] has achieved remarkable success by enabling agents to learn through interactions with their environment, using reward signals to shape their behaviour. Yet, the reward function that produces these signals is typically treated as a black box that the agent queries to receive rewards [6].

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

Expressive Reward Synthesis with the Runtime Monitoring Language

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Daniel Donnelly,
  • Francesco Belardinelli

摘要

Reinforcement Learning (RL) [10] has achieved remarkable success by enabling agents to learn through interactions with their environment, using reward signals to shape their behaviour. Yet, the reward function that produces these signals is typically treated as a black box that the agent queries to receive rewards [6].