US 2020/0034710 A1 · OPTIMIZING NEURAL NETWORK STRUCTURES FOR EMBEDDED SYSTEMS

Abstract

A model training and implementation pipeline trains models for individual embedded systems. The pipeline iterates through multiple models and estimates the performance of the models. During a model generation stage, the pipeline translates the description of the model together with the model parameters into an intermediate representation in a language that is compatible with a virtual machine. The intermediate representation is agnostic or independent to the configuration of the target platform. During a model performance estimation stage, the pipeline evaluates the performance of the models without training the models. Based on the analysis of the performance of the untrained models, a subset of models is selected. The selected models are then trained and the performance of the trained models are analyzed. Based on the analysis of the performance of the trained models, a single model is selected for deployment to the target platform.

Claims20

Dependent claims

A method for generating a machine-learned model comprising:

generating an untrained model;
generating an intermediate representation of the untrained model, the intermediate representation in an intermediate language compatible with a virtual machine;
evaluating the performance of the untrained model, wherein evaluating the performance includes at least one of determining a latency in applying the model in a target system, determining a frequency at which the model can be applied in the target system, determining an amount of resources used by the model, and determining an amount of power consumed by the target system using the model;
iteratively generating and evaluating new untrained models, a new model generated based on a performance of a previous model;
selecting a subset of models based on the performance of the generated models;
training the selected subset of models;
evaluating an accuracy of the trained models;
and
selecting a model based on the performance evaluation of the trained models for deployment to the target system.

6 more elements

Dependencies are inferred from the claim text.

Optimizing neural network structures for embedded systems

Life of this patent

Optimizing neural network structures for embedded systems

Life of this patent

Abstract

Claims20