Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces FastCentNN, an optimized version of the Centroid Neural Network (CentNN) that accelerates training by implementing an early centroid splitting strategy based on total centroid movement as a proxy for training entropy. The approach significantly reduces the duration of low-movement training phases while maintaining the original learning dynamics of CentNN. Experimental results demonstrate that FastCentNN achieves comparable clustering quality to CentNN while decreasing runtime by up to 16% on synthetic datasets and 5% on high-dimensional datasets.
FastCentNN cuts training time by up to 16% without sacrificing clustering quality, thanks to a novel early splitting strategy based on centroid movement.
Centroid neural network (CentNN) is an unsupervised competitive learning algorithm in which centroid splitting is triggered only after strict local stabilization, often leading to prolonged low-movement training phases before model expansion. This report proposes FastCentNN, an accelerated variant that addresses this inefficiency by introducing an early splitting strategy based on the total centroid movement per epoch, which serves as a training entropy proxy. As a result, FastCentNN reduces unnecessary reassignment epochs while preserving the original winner-loser learning dynamics. FastCentNN supports both absolute and stage-relative movement thresholds, allowing the splitting criterion to remain either fixed or adaptive throughout training. Experiments on some benchmark datasets show that FastCentNN consistently achieves clustering quality comparable to CentNN while reducing runtime by up to 16% on synthetic 2D datasets and about 5% on high-dimensional datasets. FastCentNN therefore provides a practical and efficient drop-in replacement for CentNN, retaining its online adaptive learning behavior while offering a simple and interpretable speed-stability trade-off through configurable splitting thresholds.