FGNG: A fast multi-dimensional growing neural gas implementation

• Fast implementation of the Growing Neural Gas algorithm.
• Accelerate the run time for a large number of nodes.
• The algorithm scales for any dimension of the feature space.

The Growing Neural Gas algorithm (GNG) is a well-known classification algorithm that is capable of capturing topological relationships that exist in the input data. Unfortunately, simple implementations of the GNG algorithm have time complexity O(n2)O(n2), where n   is the number of nodes in the graph. This fact makes these implementations impractical for use in production environments where large data sets are used. This paper aims to propose an optimized implementation that breaks the O(n2)O(n2) barrier and that addresses data in high-dimensional spaces without changing the GNG semantics. The experimental results show speedups of over 50 times for graphs with 200,000 nodes.