Reinforcement Learning Model Free with GLIE Monte-Carlo on Policy Update for Network Topology Discovery.
Roberto Casado-VaraMarcos SevertÁngel Martín del ReyHéctor QuintiánJosé Luís Calvo-RollePublished in: CISIS-ICEUTE (2022)
Keyphrases
- network topology
- monte carlo
- model free
- policy evaluation
- reinforcement learning
- policy iteration
- rl algorithms
- optimal policy
- temporal difference
- reinforcement learning algorithms
- ad hoc networks
- function approximation
- average reward
- markov chain
- network structure
- monte carlo simulation
- state space
- importance sampling
- action selection
- routing protocol
- variance reduction
- markov decision problems
- monte carlo tree search
- protein protein interactions
- stochastic approximation
- particle filter
- reward function
- policy gradient
- function approximators
- real time
- markov decision processes
- markov decision process
- partially observable markov decision processes
- partially observable
- least squares
- data mining
- decision problems
- reinforcement learning methods
- learning algorithm
- finite state