Abhishek Naik on Continuing RL & Average Reward
TalkRL: The Reinforcement Learning Podcast
00:00
Discounted vs. Average Reward Algorithms
This chapter examines the debate surrounding average and discounted reward algorithms in reinforcement learning, emphasizing the role of the discount factor gamma. It discusses the practical advantages of adapting the discount factor according to learning confidence, and the importance of separating average rewards from discount scaling for improved decision-making. Additionally, the episode reflects on personal experiences in a PhD program, highlighting the mentorship that fosters growth in research and collaboration.
Transcript
Play full episode
