Sanjeev de recoz

These are the papers:

• Ronald Parr and Lagoudakis - Least Squares Policy Iteration

• Sridhar Mahadevan - Representation Policy Iteration

• Sridhar Mahadevan - Proto-Value Functions: Developmental Reinforcement Learning.

• Bardtke and Barto - Least - Squares Temporal Difference Learning (LSTD)

• Also please note, there is a presentation, I don't remember who wrote that, its tough to search it (for you) if I find that again that on net I will mail it to you. It clearly explains Q and SARSA algorithms with eligibility traces for value function approximation