Title :
Parallel learning on the ArMenX machine by defining sub-networks
Author :
Autret, Y. ; Thepaut, A. ; Ouvradou, G. ; Le Drezen, J. ; Laisne, J.D.
Author_Institution :
Fac. des Sci., Univ. de Bretagne Occidentale, Brest, France
Abstract :
In this paper, we describe a parallel machine and show how it can be used for implementing parallel learning. The machine, called ArMenX, includes digital signal processors (DSPs) and transputers. DSPs are used for neural network computations. Transputers are used for DSP dynamic allocation. An existing learning algorithm has been chosen and implemented on the machine.
Keywords :
digital signal processing chips; learning (artificial intelligence); learning systems; neural net architecture; neural nets; parallel architectures; parallel machines; transputers; ArMenX machine; digital signal processors; learning algorithm; neural network; parallel learning machine; transputers; Computer networks; Computer vision; Digital signal processing; Field programmable gate arrays; Machine learning; Neural networks; Parallel machines; Random access memory; Read-write memory; Speech recognition;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714060
