Method for training a neural network

US6968327B1 · kind B1 · utility

Inventors

• Ronald Kates · Otterfing, DE
• Nadia Harbeck · Otterfing, DE
• Manfred Schmitt · Munich, DE

Key dates

• Filing date: Aug 24, 2000
• Grant date: Nov 22, 2005
• Expiry date: Jan 28, 2021

Classification

• Technology area (CPC G): Physics
• CPC primary: G16H50/70
• WIPO field: Medical technology
• WIPO sector: Instruments

Abstract

A method for training a neural network in order to optimize the structure of the neural network includes identifying and eliminating synapses that have no significant influence on the curve of the risk function. First and second sending neurons are selected that are connected to the same receiving neuron by respective first and second synapses. It is assumed that there is a correlation of response signals from the first and second sending neurons to the same receiving neuron. The first synapse is interrupted and a weight of the second synapse is adapted in its place. The output signals of the changed neural network are compared with the output signals of the unchanged neural network. If the comparison result does not exceed a predetermined level, the first synapse is eliminated, thereby simplifying the structure of the neural network.