Here's one way to reduce them Image classification algorithms are notoriously error-prone, but a novel method for spotting errors within incomprehensible AI code could help solve the problem.

Dr. James McCaffrey of Microsoft Research demonstrates how to fetch and prepare MNIST data for image recognition machine learning problems.

Automatic image recognition is widely used today: There are computer programs that can reliably diagnose skin cancer, navigate self-driving cars, or control robots. Up to now, all this has been ...