A Hypothesis - driven Constructive Induction Approach I ' 7 - - ' - I to Expanding Neural Networks

With most machine learning methods, if the given knowledge representation space is inadequate then the learning pnx;ess will fail. This is also true with mt:thods using neural networks as the form of the representation space. To overcome this limitation, an automatic construction method for a neural network is proposed. This paper describes the BP·HCI method for a hypothesis-driven constructive induction in a neural network trained by the backpropagation algorithm. 1be method searches for a better representation space by analyzing the hypotheses generated in each step of an iterative learning process. The method was applied to teD problems, which include, in particular. exclusive­ or, MONK2. parity-6BIT and inverse parity-6BIT problems. All problems were successfully solved with the same initial set of parameters; the extension of representation space was no more than necessary extension for each pnblem.