A Novel Graph-Based Estimation of the Distribution Algorithm and its Extension Using Reinforcement Learning

Continuous probabilistic model building genetic network programming using reinforcement learning

“…To calculate the probabilities of node connections, various approaches proposed in PMBGNP [14,16,24] can be applied. In this paper, we use a RL [33] based method called reinforced PMBGNP (RPMBGNP) [16] to construct P nc .…”

“…In this paper, we use a RL [33] based method called reinforced PMBGNP (RPMBGNP) [16] to construct P nc . In this method, the episodes of RL can be obtained during the execution of individuals: Definition 1 ((Episode).).…”

“…A novel EDA, called probabilistic model building genetic network programming (PMBGNP), was recently proposed [14][15][16]. PMBGNP is inspired by the classical EDAs, however, a distinguished directed graph (network) structure [17][18][19][20][21] is used to represent its individual.…”

“…2. Due to the unique features of its graph structures, PMBGNP explores the applicability of EDAs to wider range of problems, such as data mining [22,14,23] and the problems of controlling the agents' behavior (agent control problems) [16,[24][25][26].…”

Interoperable multi-agent framework for unmanned aerial/ground vehicles: towards robot autonomy

“…The inherent modular and distributed nature of the mult-agent system offers scalability, fault tolerance and parallelism [21,36]. Behaviour-based multi-agent architectures improve the rapid online decision-making ability of a robot by decomposing a system into subsystems with taskachieving behaviours [22,39]. Multi-agent architectures for service-oriented unmanned vehicle applications are generally platform oriented, leading to varied architectures for the same applications [13][14][15]31,33,40].…”