Delay optimal policies offer very little privacy.
Sachin KadloorNegar KiyavashPublished in: INFOCOM (2013)
Keyphrases
- optimal policy
- markov decision processes
- decision problems
- dynamic programming
- state space
- finite horizon
- reinforcement learning
- long run
- infinite horizon
- privacy preserving
- finite state
- state dependent
- sufficient conditions
- average cost
- average reward reinforcement learning
- dynamic programming algorithms
- control policies
- serial inventory systems
- markov decision process
- initial state
- multistage
- average reward
- policy iteration
- total reward
- inventory level
- markov decision problems
- bayesian reinforcement learning
- partially observable markov decision processes