Publication Details
Self-Organized Sparse Distributed Memory -- an Application
Neural Net, Self-Organizing Map, Soft Competitive Learning Rule, Sparse Distributed Memory, Prediction
The paper discuses an extension of Kanerva's Sparse Distributed Memory (SDM) and introduces possible application in storage and retrieval sequences (predicting). A self-organizing SDM equivalent to a three-layered neural network is used to learn the desired transfer function mapping a sequence (history) of 2D points to next point. The net was tested in simple experiment, where some kinds of Lissajous curves was successfully stored and retrieved.
The paper discuses an extension of Kanerva's Sparse Distributed Memory (SDM) and introduces possible application in storage and retrieval sequences (predicting). A self-organizing SDM equivalent to a three-layered neural network is used to learn the desired transfer function mapping a sequence (history) of 2D points to next point. The net was tested in simple experiment, where some kinds of Lissajous curves was successfully stored and retrieved.
@INPROCEEDINGS{FITPUB5788, author = "Franti\v{s}ek Greben\'{i}\v{c}ek", title = "Self-Organized Sparse Distributed Memory -- an Application", pages = "39--44", booktitle = "ASIS '99", year = 1999, location = "Krnov, CZ", ISBN = "80-85988-41-0", language = "english", url = "https://www.fit.vut.cz/research/publication/5788" }