0

I'm working on a research project that involves the application of reinforcement learning to planning and decision-making problems. Typically, these problems involve picking (sampling) multiple actions within a state based on ranking [max_q to min_q]. The RL literature seems focused on policies that map from a set of states to a single individual action not multiple. Does anyone know of approaches that can not only map states to multiple simultaneous actions but also that can maintain relationships between these actions? Here is the DQN action selection function source code

Thanks

0 Answers0