Q-Learning
Q-Learning
Q-Learning
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Q <strong>Learning</strong> Algorithm<br />
For each state-action pair (s, a), initialize the table entry Q s , a to zero<br />
Observe the current state s<br />
Do forever:<br />
---Select an action a and execute it<br />
---Receive immediate reward r<br />
---Observe the new state s'<br />
---Update the table entry for Qs , a as follows:<br />
Qs ,a=r max a ' Q s' ,a ' <br />
--- s=s '