"Thoughts on machine learning accuracy," 27 Jul 2018

This blog shares some brief thoughts on machine learning accuracy and bias...  Using Rekognition, the ACLU built a face database using 25,000 publicly available arrest photos and then performed facial similarity searches on that database using public photos of all current members of Congress. They found 28 incorrect matches out of 535... Some thoughts on their claims:

• The default confidence threshold for facial recognition APIs in Rekognition is 80%, which is good for a broad set of general use cases... but it’s not the right setting for public safety use cases... We recommend 99% for use cases where highly accurate face similarity matches are important...
• In real-world public safety and law enforcement scenarios, Amazon Rekognition is almost exclusively used to help narrow the field and allow humans to expeditiously review and consider options using their judgment...,where it can help find lost children, fight against human trafficking, or prevent crimes. 

There’s a difference between using machine learning to identify a food object and using machine learning to determine whether a face match should warrant considering any law enforcement action. The latter is serious business and requires much higher confidence levels. We continue to recommend that customers do not use less than 99% confidence levels for law enforcement matches, and then to only use the matches as one input across others that make sense for each agency.