概要:

The learning ability and the memory ability of the human brain have the excellent processing performance that the present computers can't achieve. The neural network is a technology that reproduces the human brain artificially.

The hierarchical neural network that has the internal memory layer is suitable for learning of time-series. The conventional back-propagetion was not effective for learning the network. And so, the ``delayed learning method'' was newly devised. Using this method, the neural network with internal memory can learn complicated time-series. But, the parameters used for learning was decided by experience. In this study, the influence on the learning result of ``delay time'' is examined. ``Delay time'' is one of the parameters of the delayed learning method. The network composed of several elements tried learning many kinds of time-series. And the difference of a learning result of each delay time was compared and discussed. The experimental results showed that the number of data per period has a big influence on deciding effective delay time. Changing other parameters showed the same result.