Global collaboration through local interaction in competitive learning.

Journal: Neural networks : the official journal of the International Neural Network Society
Published Date:

Abstract

Feature maps, that preserve the global topology of arbitrary datasets, can be formed by self-organizing competing agents. So far, it has been presumed that global interaction of agents is necessary for this process. We establish that this is not the case, and that global topology can be uncovered through strictly local interactions. Enforcing uniformity of map quality across all agents results in an algorithm that is able to consistently uncover the global topology of diversely challenging datasets. The applicability and scalability of this approach is further tested on a large point cloud dataset, revealing a linear relation between map training time and size. The presented work not only reduces algorithmic complexity but also constitutes first step towards a distributed self organizing map.

Authors

  • Abbas Siddiqui
    Future Resilient Systems Singapore-ETH Centre, 1 Create Way CREATE Tower, Singapore; ETH Zurich, 8092 Zurich, Switzerland. Electronic address: abbas.siddiqui@frs.ethz.ch.
  • Dionysios Georgiadis
    Future Resilient Systems Singapore-ETH Centre, 1 Create Way CREATE Tower, Singapore; ETH Zurich, 8092 Zurich, Switzerland. Electronic address: dionysios.georgiadis@frs.ethz.ch.