Winner-take-all networks: Time-based versus activation-based mechanisms for various selection goals

A time-based temporal-winner-take-all (TWTA) mechanism, with O (n) space complexity and roughly O(log n) time complexity is presented. It exploits systematic and stochastic differences between time delays within different units and connections. TWTA is logically equivalent to standard winner-take-all (WTA), but TWTA may be more suitable for various selection tasks, especially the selection of an arbitrary unit from a set (e.g. as in unit recruitment). TWTA avoids various problems with conventional WTA, notably the difficulty of making it converge rapidly over a large range of conditions. A mathematical analysis of the mechanism and simulations of it are reported. TWTA and WTA are compared and discussed