Title :
Unsupervised learning algorithm for signal separation
Author :
Jacob, Theju ; Snyder, Wesley
Author_Institution :
Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA
Abstract :
We present a neural network capable of separating inputs in an unsupervised manner. Oja´s rule and Self-Organizing map principles are used to construct the network. The network is tested using 1) straight lines 2)MNIST database. The results demonstrate that the network can operate as a general clustering algorithm, with neighboring neurons responding to geometrically similar inputs.
Keywords :
pattern clustering; self-organising feature maps; source separation; unsupervised learning; Oja rule; general clustering algorithm; neural network; neurons; self-organizing map principles; signal separation; unsupervised learning algorithm; Clustering algorithms; Hebbian theory; Lattices; Neurons; Training;
Conference_Titel :
Computational Intelligence for Multimedia, Signal and Vision Processing (CIMSIVP), 2014 IEEE Symposium on
Conference_Location :
Orlando, FL
DOI :
10.1109/CIMSIVP.2014.7013286
