Rohin Shah
TalkRL: The Reinforcement Learning Podcast
00:00
The Relationship Between Inverse Reenforcement Learning and Reward Learning
In inverse reinforcement learning, you try to cover the reward function fom observing an agent's behavior. We're just saying, nanana, we're not watching the human's behavior. And so you can think of this in the framework of inverse reenforcement learning. So with this approach, i guess potentially you're opening up a huge amount of kind of learning just from observing what's happening. But yes, if you had a enough data such that sall, supervised learning could teach the agent a bunch about dynamics.
Transcript
Play full episode
