DETECTING SYSTEMATIC STRUCTURE IN DISTRIBUTED REPRESENTATIONS

A novel method of analysing the systematic structure formed inside a f eedforward neural network is developed. This method is applied to the task of understanding how a network is performing unification on distr ibuted representations. Systematic structure is detected by examining inter-representational distances, rather than by attempting to detect the vectorial similarities sought by previously described techniques. The method described can be applied to the detection of systematic str ucture in other networks using distributed representations. (C) 1998 E lsevier Science Ltd. All rights reserved.