Optimizing expectation with guarantees in POMDPs

• المؤلفون الرئيسيون: CHATTERJEE, Krishnendu, PEREZ, Guillermo A., RASKIN, Jean-François, ZIKELIC, Dorde
• التنسيق: text
• اللغة: English
• منشور في: Institutional Knowledge at Singapore Management University 2017
• الموضوعات: Artificial Intelligence and Robotics
• الوصول للمادة أونلاين: https://ink.library.smu.edu.sg/sis_research/9071; https://ink.library.smu.edu.sg/context/sis_research/article/10074/viewcontent/11046_13_14574_1_2_20201228.pdf
• المؤسسة: Singapore Management University

A standard objective in partially-observable Markov decision processes (POMDPs) is to find a policy that maximizes the expected discounted-sum payoff. However, such policies may still permit unlikely but highly undesirable outcomes, which is problematic especially in safety-critical applications. Re...