Temporal Graph Traversals Using Reinforcement Learning With Proximal Policy Optimization.
Samuel Henrique SilvaAdel AlaeddiniPeyman NajafiradPublished in: IEEE Access (2020)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- action selection
- state space
- optimization algorithm
- markov decision processes
- spatial and temporal
- function approximation
- optimal control
- temporal constraints
- control policy
- learning algorithm
- graph structure
- optimization problems
- spatio temporal
- dynamic programming
- temporal difference
- agent receives
- partially observable environments
- reinforcement learning problems
- objective function
- policy gradient
- policy iteration
- markov decision process
- graph representation
- graph model
- temporal information
- infinite horizon
- graph matching
- reward function
- reinforcement learning algorithms
- temporal databases
- function approximators
- directed acyclic graph
- approximate dynamic programming
- random walk
- directed graph
- temporal reasoning