A new approach to intelligent decision-making in reinforcement learning

An agent can be understood in terms of 3 parts working together: the knowledge part, the reasoning part, and the goal. The knowledge part retains everything the agent has learned from its entire history. The reasoning part is a computational process that processes newly arrived information based on the learned knowledge. It helps condense information into various kinds of knowledge and concludes the actions the agent should take. The goal is something that orients the reasoning process and is aligned with the agent’s lifelong interests. Credit: Chenyang Wu et al.

A new paper published in Intelligent Computing presents the primary challenges …
Read more…….

Be the first to comment

Leave a Reply

Your email address will not be published.


*