Approximating Optimal Policies for Agents with Limited Execution Resources.
Dmitri A. DolgovEdmund H. DurfeePublished in: IJCAI (2003)
Keyphrases
- optimal policy
- markov decision processes
- resource allocation
- multi agent
- multi agent systems
- finite horizon
- reinforcement learning
- state space
- decision problems
- long run
- dynamic programming
- multiple agents
- finite state
- expected reward
- infinite horizon
- multistage
- average reward
- dynamic programming algorithms
- state dependent
- serial inventory systems
- policy iteration
- decision making
- sufficient conditions
- initial state
- single agent
- average cost
- decision theoretic
- markov decision process
- action selection
- software agents
- average reward reinforcement learning
- inventory level
- finite number
- dynamic environments