Explores approximation algorithms for resource allocation and network design problems, competitive algorithms for the TCP acknowledgment problem, and experimental results.
Covers model-free prediction methods in reinforcement learning, focusing on Monte Carlo and Temporal Differences for estimating value functions without transition dynamics knowledge.
