Assume the goal is to train an agent set in any room to go outside. 5...

 

Transcribed Image Text:

Assume the goal is to train an agent set in any room to go outside. 5 Finish Start Depict this as reinforcement learning problem, with a graph where nodes are rooms and arcs are actions labeled with rewards. Compute Q (1; 1 5) and then Q (3; 3 1) assuming Y=0.9 Ac Go Assume the goal is to train an agent set in any room to go outside. 5 Finish Start Depict this as reinforcement learning problem, with a graph where nodes are rooms and arcs are actions labeled with rewards. Compute Q (1; 1 5) and then Q (3; 3 1) assuming Y=0.9 Ac Go