A Parameter-free Adaptive Resonance Theory-based Topological Clustering Algorithm Capable of Continual Learning

TL;DR

An ART-based topological clustering algorithm that integrates parameter estimation methods for both the similarity threshold and the edge deletion threshold is proposed and has superior clustering performance to state-of-the-art clustering algorithms without requiring parameter specifications specific to the datasets.

Abstract

In general, a similarity threshold (i.e., a vigilance parameter) for a node learning process in Adaptive Resonance Theory (ART)-based algorithms has a significant impact on clustering performance. In addition, an edge deletion threshold in a topological clustering algorithm plays an important role in adaptively generating well-separated clusters during a self-organizing process. In this paper, we propose an ART-based topological clustering algorithm that integrates parameter estimation methods for both the similarity threshold and the edge deletion threshold. The similarity threshold is estimated using a determinantal point process-based criterion, while the edge deletion threshold is defined based on the age of edges. Experimental results with synthetic and real-world datasets show that the proposed algorithm has superior clustering performance to state-of-the-art clustering algorithms without requiring parameter specifications specific to the datasets. Source code is available at https://github.com/Masuyama-lab/CAE

Figures (17)

• Figure 1: Flowchart of the CAE learning procedure. The procedure is identical to the CAEA learning process except for the parameter estimation steps highlighted in red.
• Figure 2: Two-dimensional synthetic dataset.
• Figure 3: Visualization of the synthetic dataset in sequential order (i.e., (a) to (f)).
• Figure 4: Visualization of self-organizing results in the stationary environment.
• Figure 5: Visualization of self-organizing results of AutoCloud in the non-stationary environment.