FAST NON-NEGATIVE LEAST-SQUARES LEARNING IN THE RANDOM NEURAL NETWORK. Issue 3 (18th May 2016)

• Record Type: Journal Article
• Authors: Timotheou, Stelios
• Is Part Of: Probability in the engineering and informational sciences. Volume 30:Issue 3(2016)
• Page Start: 379
• Page End: 402
• Publication Date: 2016-05-18
• Subjects: Probabilities -- Periodicals; Engineering -- Statistical methods -- Periodicals; Information science -- Statistical methods -- Periodicals; 519.202462
• DOI: 10.1017/S0269964816000061
• Languages: English
• ISSNs: 0269-9648

Abstract

Abstract : The random neural network is a biologically inspired neural model where neurons interact by probabilistically exchanging positive and negative unit-amplitude signals that has superior learning capabilities compared to other artificial neural networks. This paper considers non-negative least squares supervised learning in this context, and develops an approach that achieves fast execution and excellent learning capacity. This speedup is a result of significant enhancements in the solution of the non-negative least-squares problem which regard (a) the development of analytical expressions for the evaluation of the gradient and objective functions and (b) a novel limited-memory quasi-Newton solution algorithm. Simulation results in the context of optimizing the performance of a disaster management problem using supervised learning verify the efficiency of the approach, achieving two orders of magnitude execution speedup and improved solution quality compared to state-of-the-art algorithms.