Clustering for Improved Learning in Maze Traversal Problem

TL;DR

This paper enhances maze traversal learning by clustering maze features and modifying the CSRN to incorporate external inputs, resulting in faster and more effective training.

Contribution

It introduces a clustering approach and modifies the CSRN with external inputs to improve maze traversal learning efficiency.

Findings

Clustering maze features improves CSRN learning.

External inputs enhance training speed.

Modified CSRN outperforms previous models.

Abstract

The maze traversal problem (finding the shortest distance to the goal from any position in a maze) has been an interesting challenge in computational intelligence. Recent work has shown that the cellular simultaneous recurrent neural network (CSRN) can solve this problem for simple mazes. This thesis focuses on exploiting relevant information about the maze to improve learning and decrease the training time for the CSRN to solve mazes. Appropriate variables are identified to create useful clusters using relevant information. The CSRN was next modified to allow for an additional external input. With this additional input, several methods were tested and results show that clustering the mazes improves the overall learning of the traversal problem for the CSRN.