MATCH TRACKING STRATEGIES FOR FUZZY ARTMAP NEURAL NETWORKS

Training fuzzy ARTMAP neural networks for classification using data from complex real-world environments may lead to category proliferation, and yield poor performance. This problem is known to occur whenever the training set contains noisy and overlapping data. Moreover, when the training set contains identical input patterns that belong to different recognition classes, fuzzy ARTMAP will fail to converge. To circumvent these problems, some alternatives to the network’s original match tracking (MT) process have been proposed in literature, such as using negative MT, and removing MT altogether. In this chapter, the MT parameter of fuzzy ARTMAP is optimized during training using a new Particle Swarm Optimisation (PSO)-based strategy, denoted PSO (MT). The impact on fuzzy ARTMAP performance of training with different MT strategies is assessed empirically, using different synthetic data sets, and the NIST SD19 handwritten character recognition data set. During computer simulations, fuzzy ARTMAP is trained with the original (positive) match tracking (MT+), with negative match tracking (MT−), without MT algorithm (WMT), and with PSO (MT). Through a comprehensive set of simulations, it has been observed that by training with MT−, fuzzy ARTMAP expends fewer resources than with other MT strategies, but can achieve a significantly higher generalization error, especially for data with overlapping class distributions. In particular, degradation of error in fuzzy ARTMAP performance due to overtraining is more pronounced for MT− than for MT+. Generalization error achieved using WMT is significantly higher than other strategies on data with complex non-linear decision bounds. Furthermore, the number of internal categories required to represent decision boundaries increases significantly. Optimizing the value of the match tracking parameter using PSO (MT) yields the lowest overall generalization error, and requires fewer internal categories than WMT, but generally more categories than MT+ and MT−. However, this strategy requires a large number of training epochs to convergence. Based on this empirical results with PSO (MT), the MT process as such can provide a significant increase to fuzzy ARTMAP performance, assuming that the MT parameter is tuned for the specific application in mind.