Training Humans and Machines

TL;DR

This paper reviews shared principles of learning in humans and machines, highlighting potential new applications to enhance learning speed, retention, and generalizability across both domains.

Contribution

It synthesizes existing research on learning principles and suggests novel cross-domain applications for improving learning methods.

Findings

Shared principles underpin human and machine learning

Potential for novel applications in education and AI

Highlights areas for future research

Abstract

For many years, researchers in psychology, education, statistics, and machine learning have been developing practical methods to improve learning speed, retention, and generalizability, and this work has been successful. Many of these methods are rooted in common underlying principles that seem to drive learning and overlearning in both humans and machines. I present a review of a small part of this work to point to potentially novel applications in both machine and human learning that may be worth exploring.