US 11,455,540 B2 · Autonomic horizontal exploration in neural networks transfer learning

Abstract

An autonomic function is caused to execute in an artificial intelligence environment to detect a new problem space. Using the autonomic function, a first model is selected. The first model includes a first trained neural network corresponding to a first ontology. A second model is automatically identified. the second model includes a second trained neural network corresponding to a second ontology. A layer is autonomically extracted from the second model and inserted into a location in the first model. A vector transformation is automatically constructed to transform an output vector of a previous layer in an immediately previous location in the model relative to the location. The layer is automatically fused in the first model using the transformed output vector as input to the layer, the fusing forming a fused model that is operable on an ontology of the new problem space.

Claims20

Dependent claims

A method comprising:

causing an autonomic function to execute using a processor and a memory in an artificial intelligence environment to detect a new problem space;
selecting, using the autonomic function, a first model, wherein the first model comprises a first trained neural network corresponding to a first ontology;
identifying, using the autonomic function, a second model, wherein the second model comprises a second trained neural network corresponding to a second ontology;
extracting, using the autonomic function, a layer from the second model;
inserting, using the autonomic function, the layer into a location in the first model;
constructing, using the autonomic function, a vector transformation, wherein the vector transformation transforms an output vector of a previous layer in an immediately previous location in the first model relative to the location;
and fusing, using the autonomic function, the layer in the first model using the transformed output vector as input to the layer, the fusing forming a fused model that is operable on an ontology of the new problem space.

4 more elements

A computer usable program product comprising one or more computer-readable storage medium, and program instructions collectively stored on the one or more computer-readable storage medium, the stored program instructions comprising:

program instructions to cause an autonomic function to execute using a processor and a memory in an artificial intelligence environment to detect a new problem space;
program instructions to select, using the autonomic function, a first model, wherein the first model comprises a first trained neural network corresponding to a first ontology;
program instructions to identify, using the autonomic function, a second model, wherein the second model comprises a second trained neural network corresponding to a second ontology;
program instructions to extract, using the autonomic function, a layer from the second model;
program instructions to insert, using the autonomic function, the layer into a location in the first model;
program instructions to construct, using the autonomic function, a vector transformation, wherein the vector transformation transforms an output vector of a previous layer in an immediately previous location in the first model relative to the location;
and program instructions to fuse, using the autonomic function, the layer in the first model using the transformed output vector as input to the layer, the program instructions to fuse forming a fused model that is operable on an ontology of the new problem space.

4 more elements

Dependencies are inferred from the claim text.

Autonomic horizontal exploration in neural networks transfer learning

Life of this patent

Autonomic horizontal exploration in neural networks transfer learning

Life of this patent

Abstract

Claims20