Generating and managing deep tensor neural networks

US11531902B2 · kind B2 · utility

Assignees

• International Business Machines Corporation · US
• Trustees of Tufts College · US
• Ramot at Tel-Aviv University Ltd. · IL

Inventors

• Lior Horesh · North Salem, US
• Elizabeth Newman · Rehovot, IL
• Misha E. Kilmer · Medford, US
• Haim Avron · White Plains, US

Key dates

• Filing date: Nov 13, 2018
• Grant date: Dec 20, 2022
• Expiry date: Apr 25, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/084
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Techniques for generating and managing, including simulating and training, deep tensor neural networks are presented. A deep tensor neural network comprises a graph of nodes connected via weighted edges. A network management component (NMC) extracts features from tensor-formatted input data based on tensor-formatted parameters. NMC evolves tensor-formatted input data based on a defined tensor-tensor layer evolution rule, the network generating output data based on evolution of the tensor-formatted input data. The network is activated by non-linear activation functions, wherein the weighted edges and non-linear activation functions operate, based on tensor-tensor functions, to evolve tensor-formatted input data. NMC trains the network based on tensor-formatted training data, comparing output training data output from the network to simulated output data, based on a defined loss function, to determine an update. NMC updates the network, including weight and bias parameters, based on the update, by application of tensor-tensor operations.