Calibrating Artificial Neural Networks by Global Optimization
Janos D. Pinter(janos.pinterozyegin.edu.tr)
Abstract: An artificial neural network (ANN) is a computational model - implemented as a computer program - that is aimed at emulating the key features and operations of biological neural networks. ANNs are extensively used to model unknown or unspecified functional relationships between the input and output of a “black box” system. In order to apply such a generic procedure to actual decision problems, a key requirement is ANN training to minimize the discrepancy between modeled and measured system output. In this work, we consider ANN training as a (potentially) multi-modal optimization problem. To address this issue, we introduce a global optimization (GO) framework and corresponding GO software. The practical viability of the GO based approach is illustrated by finding close numerical approximations of (one-dimensional, but non-trivial) functions.
Keywords: Artificial Neural Networks, ANN model calibration by global optimization, Lipschitz Global Optimizer (LGO) solver suite, ANN implementation in Mathematica, MathOptimizer Professional (LGO for Mathematica), Illustrative numerical examples.
Category 1: Global Optimization (Applications )
Category 2: Applications -- Science and Engineering (Data-Mining )
Category 3: Optimization Software and Modeling Systems
Citation: Technical Report, Özyeğin University, Istanbul. Submitted for publication: July 2010.
Entry Submitted: 07/21/2010
Modify/Update this entry
|Visitors||Authors||More about us||Links|
Search, Browse the Repository
Give us feedback
|Optimization Journals, Sites, Societies|