Safety through feedback in constrained RL

• Main Authors: CHIRRA, Shashank Reddy, VARAKANTHAM, Pradeep, PARUCHURI, Praveen
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: Reinforcement Learning; Cost function; Machine learning; Feedback; Artificial Intelligence and Robotics; Computer Sciences
• Online Access: https://ink.library.smu.edu.sg/sis_research/9968; https://ink.library.smu.edu.sg/context/sis_research/article/10968/viewcontent/2406.19626v22.pdf
• Institution: Singapore Management University

In safety-critical RL settings, the inclusion of an additional cost function is often favoured over the arduous task of modifying the reward function to ensure the agent's safe behaviour. However, designing or evaluating such a cost function can be prohibitively expensive. For instance, in the...