Project: #93 Incorporating Uncertainty for Reinforcement Learning of Agent Policies Progress Report - Reporting Period Ending: March 30, 2018 Principal Investigator: Daniel Lee Status: Active Start Date: None End Date: None Research Type: Basic Grant Type: Research Grant Program: FAST Act Grant Cycle: 2017 Mobility21 UTC Progress Report (Last Updated: March 30, 2018, 6:58 a.m.) % Project Completed to Date: 33 % Grant Award Expended: 0 % Match Expended & Document: 0 USDOT Requirements Accomplishments We have introduced a new Bayesian approach to Reinforcement Learning using off-policy TD methods and Assumed Density Filtering. This allows for updates on action-values (Q) through an online Bayesian inference method. Our empirical results show our proposed algorithms outperform comparable algorithms in several task domains. Impacts The current work will enable more efficient training of autonomous systems for intelligent transportation systems. In particular, it will allow for better exploration of policies in decision making processes. This will be critical in systems that need to navigate in uncertain traffic environments. Other Open source software available through Penn. New Partners N/A Issues N/A