A Submodule in Rainbow DQN: Multi-Step Learning
1. What is a “Merged Experience Tuple”? A single-step experience tuple is the data generated from each interaction between the agent and the environment<span>(s, a, r, s', terminal, done)</span>, while a “merged experience tuple” combines N consecutive single-step experiences into a “multi-step experience tuple”. The purpose is to use the “N-step cumulative reward” instead of … Read more