We investigate attractor neural networks with a modular structure, where a local winner-takes-all rule acts within the modules (called hyper-columns). We make a signal-to-noise analysis of storage capacity and noise tolerance, and compare the results with those from simulations. Introducing local winner-takes-all dynamics improves storage capacity and noise tolerance, while the optimal size of the hypercolumns depends on network size and noise level. © Springer-Verlag Berlin Heidelberg 2002.
CITATION STYLE
Johansson, C., Sandberg, A., & Lansner, A. (2002). Attractor neural networks with hypercolumns. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2415 LNCS, pp. 192–197). Springer Verlag. https://doi.org/10.1007/3-540-46084-5_32
Mendeley helps you to discover research relevant for your work.