Scholarlink

We are presenting a proof-of-concept means of training decision supportDecision support agents to assist in optimally selecting design synthesis activities in a way that incorporates balanced risk avoidance, risk acceptance, and risk mitigation, including the costs (and value) of necessary verificationVerification activities. The proof-of-concept simulates synthesis of design specifications using a partially observed Markov decision processPartially observed markov decision process (POMDP). These simulationsSimulation are used to train decision supportDecision support agents by means of reinforcement learning (RL) and approximate dynamic programming (ADP). Trained agents may be evaluated by analyzing their policy preferences given an observation of the design synthesis state.

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

Engineering Design Synthesis Simulator: A POMDP Approach to Training Decision Support Agents for Design Synthesis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Luke Florer,
  • Rajesh Ganesan

摘要

We are presenting a proof-of-concept means of training decision supportDecision support agents to assist in optimally selecting design synthesis activities in a way that incorporates balanced risk avoidance, risk acceptance, and risk mitigation, including the costs (and value) of necessary verificationVerification activities. The proof-of-concept simulates synthesis of design specifications using a partially observed Markov decision processPartially observed markov decision process (POMDP). These simulationsSimulation are used to train decision supportDecision support agents by means of reinforcement learning (RL) and approximate dynamic programming (ADP). Trained agents may be evaluated by analyzing their policy preferences given an observation of the design synthesis state.