Online Linear Subspace Learning in an Analog Array Computing Architecture
Conference: CNNA 2016 - 15th International Workshop on Cellular Nanoscale Networks and their Applications
08/23/2016 - 08/25/2016 at Dresden, Deutschland
Proceedings: CNNA 2016
Pages: 2Language: englishTyp: PDF
Personal VDE Members are entitled to a 10% discount on this title
Authors:
Poikonen, Jussi H.; Laiho, Mika (Technology Research Center, University of Turku, Finland)
Abstract:
We map to an analog array computing framework a recently presented algorithm for online learning of linear subspaces using local learning rules. We demonstrate that the considered algorithm is well suited for implementation in analog array computing architectures, and that the computation of neural activity dynamics required in the learning can be realized in an efficient homeostatic manner using an analog continuoustime circuit.