Local search and pseudoinversion: an hybrid approach to neural network training

• Published in: Knowledge and information systems Vol. 48; no. 2; pp. 493 - 503
• Main Authors: Rubini, Luca, Cancelliere, Rossella, Gallinari, Patrick, Grosso, Andrea
• Format: Journal Article
• Language: English
• Published: London Springer London 01-08-2016; Springer Nature B.V
• Subjects: Algorithms; Analysis; Artificial intelligence; Back propagation; Computer Science; Data Mining and Knowledge Discovery; Database Management; Generalized inverse; Impact analysis; Information Storage and Retrieval; Information systems; Information Systems and Communication Service; Information Systems Applications (incl.Internet); IT in Business; Mathematical analysis; Neighborhoods; Neural networks; Random sampling; Regularization methods; Searching; Short Paper; Studies; Training; Random projections; Local search; Neural networks; Pseudoinverse matrix
• ISSN: 0219-1377; 0219-3116
• DOI: 10.1007/s10115-016-0935-y

We consider recent successful techniques proposed for neural network training that set randomly the weights from input to hidden layer, while weights from hidden to output layer are analytically determined by Moore–Penrose generalized inverse. This study aimed to analyse the impact on performances when the completely random sampling of the space of input weights is replaced by a local search procedure over a discretized set of weights. The performances of the proposed training methods are assessed through computational experience on several UCI datasets.