Approximation Algorithms for Partial-Information Based Stochastic Control with Markovian Rewards.
Sudipto GuhaKamesh MunagalaPublished in: FOCS (2007)
Keyphrases
- approximation algorithms
- partial information
- stochastic control
- control problems
- reinforcement learning
- queueing systems
- optimal control
- incomplete information
- np hard
- operations management
- special case
- brownian motion
- markov decision processes
- worst case
- constant factor approximation
- queueing networks
- reward function
- partial differential equations
- scheduling problem
- multi agent
- planning domains
- lower bound