Relevance score assignment for artificial neural networks

US12061966B2 · kind B2 · utility

Assignees

• Fraunhofer-Gesellschaft zur förderung der angewandten Forschung e.V. · DE
• TECHNISCHE UNIVERSITÄT BERLIN · DE

Inventors

• Sebastian LAPUSCHKIN · Berlin, DE
• Wojciech SAMEK · Berlin, DE
• Klaus-Robert Mueller · Berlin, DE
• Alexander K. Binder · Findlay, US
• Grégoire Montavon · Berlin, DE

Key dates

• Filing date: Sep 20, 2017
• Grant date: Aug 13, 2024
• Expiry date: Jun 15, 2043

Classification

• Technology area (CPC G): Physics
• CPC primary: G10L25/30
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

The task of relevance score assignment to a set of items onto which an artificial neural network is applied is obtained by redistributing an initial relevance score derived from the network output, onto the set of items by reversely propagating the initial relevance score through the artificial neural network so as to obtain a relevance score for each item. In particular, this reverse propagation is applicable to a broader set of artificial neural networks and/or at lower computational efforts by performing same in a manner so that for each neuron, preliminarily redistributed relevance scores of a set of downstream neighbor neurons of the respective neuron are distributed on a set of upstream neighbor neurons of the respective neuron according to a distribution function.