Parallel neural learning for control problems on a bus-based architecture

In [6], we distributed training instances over a single-channel broadcast c ommunication model to speed up execution of the back-propagation learning a lgorithm for classification problems. In this paper, we extend this concept to control problems, where the output is not necessarily 0 or 1, but range s over an interval. We first propose a modified back-propagation learning a lgorithm that incrementally decreases the error threshold by half in order to process training instances with large weight changes as quickly as possi ble. This modified back-propagation learning algorithm is then parallelized using the single-channel broadcast communication model to n processors, wh ere n is the number of training instances. Finally, the parallel back-propa gation learning algorithm is modified for execution on a bounded number of processors to cope with real-world conditions.