Curiosity learning

Up until now, we have only ever considered external rewards given to the agent from the environment. Yet, we and other animals receive a wide variety of external and internal rewards. Internal rewards are often characterized by emotion or feeling. An agent could have an internal reward that gives it +1 every time it looks to some face, perhaps denoting some internal love or infatuation reward. These types of rewards are called intrinsic rewards and they represent rewards that are internal or self-derived by the agent. This has some powerful capabilities for everything from creating interesting motivated agents to enhancing an agent's learning ability.

It is the second way in which Unity introduced curiosity learning or ...

Get Hands-On Reinforcement Learning for Games now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.