Skip to content

NEAT-QL

Latest
Compare
Choose a tag to compare
@kholoshnia kholoshnia released this 09 Dec 19:04
· 13 commits to master since this release
2292810

Neuroevolution of augmenting topologies, Q-Learning

Universal and adaptable algorithm for finding the shortest path based on a self-learning neural network and its software implementation was developed. The universality of the algorithm lies in the fact that the User can independently determine the script for the execution of the algorithm - choose the optimal learning algorithm, create the necessary map of the area and set the logic of the movement of the object. To demonstrate these features, a user interface was developed for an effective software implementation of the algorithm. An algorithm for visualizing the learning process of the genetic algorithm and reinforcement learning for finding the shortest path has been developed.