{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/self-improving-safety-performance-of","title":"Self-Improving Safety Performance of Reinforcement Learning Based Driving with Black-Box Verification Algorithms","arxiv_id":"2210.16575","date":"2022-10-29","proceeding":null,"authors":["Resul Dagdanov","Halil Durmus","Nazim Kemal Ure"],"abstract":"In this work, we propose a self-improving artificial intelligence system to enhance the safety performance of reinforcement learning (RL)-based autonomous driving (AD) agents using black-box verification methods. RL algorithms have become popular in AD applications in recent years. However, the performance of existing RL algorithms heavily depends on the diversity of training scenarios. A lack of safety-critical scenarios during the training phase could result in poor generalization performance in real-world driving applications. We propose a novel framework in which the weaknesses of the training set are explored through black-box verification methods. After discovering AD failure scenarios, the RL agent's training is re-initiated via transfer learning to improve the performance of previously unsafe scenarios. Simulation results demonstrate that our approach efficiently discovers safety failures of action decisions in RL-based adaptive cruise control (ACC) applications and significantly reduces the number of vehicle collisions through iterative applications of our method. The source code is publicly available at https://github.com/data-and-decision-lab/self-improving-RL.","url_abs":"https://arxiv.org/abs/2210.16575v3","url_pdf":"https://arxiv.org/pdf/2210.16575v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"self-improving-safety-performance-of","repo_url":"https://github.com/data-and-decision-lab/self-improving-RL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"self-improving-safety-performance-of","repo_url":"https://github.com/resuldagdanov/self-improving-RL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"safe-reinforcement-learning","task_name":"Safe Reinforcement Learning"},{"task_slug":"self-learning","task_name":"Self-Learning"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"discriminative-adversarial-search","method_name":"Discriminative Adversarial Search"},{"method_slug":"entropy-regularization","method_name":"Entropy Regularization"},{"method_slug":"monte-carlo-tree-search","method_name":"Monte-Carlo Tree Search"},{"method_slug":"ppo","method_name":"PPO"},{"method_slug":"selective-search","method_name":"Selective Search"},{"method_slug":"self-adversarial-negative-sampling","method_name":"Self-Adversarial Negative Sampling"},{"method_slug":"self-adaptive-training","method_name":"Self-adaptive Training"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}