Training a machine learning model with limited training data

A computer-implemented method, comprising:

• receiving, at a client device, a trained machine learning model that has been transformed by replacing at least one layer of the trained machine learning model with a dictionary matrix and a coefficient matrix, the dictionary matrix and the coefficient matrix formed by at least decomposing a weight matrix associated with the at least one layer of the trained machine learning model, and a product of the dictionary matrix and the coefficient matrix comprising a reduced-dimension representation of the weight matrix associated with the at least one layer of the trained machine learning model;
• updating, at the client device and based at least on a first training data acquired at the client device, the trained machine learning model;
• and
• applying, at the client device, the updated machine learning model to perform one or more tasks.

A non-transitory computer-readable medium storing instructions, which when executed by at least one data processor, result in operations comprising:

• receiving, at a client device, a trained machine learning model that has been transformed by replacing at least one layer of the trained machine learning model with a dictionary matrix and a coefficient matrix, the dictionary matrix and the coefficient matrix formed by at least decomposing a weight matrix associated with the at least one layer of the trained machine learning model, and a product of the dictionary matrix and the coefficient matrix comprising a reduced-dimension representation of the weight matrix associated with the at least one layer of the trained machine learning model;
• updating, at the client device and based at least on a first training data acquired at the client device, the trained machine learning model;
• and
• applying, at the client device, the updated machine learning model to perform one or more tasks.