Model compression by sparsity-inducing regularization optimization

A computer-implemented method comprising:

• receiving a machine learning model and parameters for the machine learning model;
• updating a next iteration of the machine learning model based on parameters for a current iteration of the machine learning model, wherein the machine learning model is updated using a sparsity-inducing regularization optimization process;
• and
• executing the updated machine learning model to generate one or more outputs.

A computing system, comprising:

• one or more processors;
• and
• a computer-readable storage medium having computer-executable instructions stored thereupon which, when executed by the processor, cause the computing system to perform operations comprising:
• receiving a neural network and parameters for the neural network;
• updating a next iteration of the neural network based on parameters for a current iteration of the neural network, wherein the neural network is updated using a sparsity-inducing regularization optimization process;
• and
• executing the updated neural network to generate one or more outputs.