Unforeseen data may be presented to the computer vision system during operation despite careful mitigation of datasets and shortcuts.

Nobel Prize-winning economist, Daniel Kahneman once remarked “by their very nature, heuristic shortcuts will produce biases, and that is true for both humans and artificial intelligence, but their heuristics of AI are not necessarily the human ones”. This is certainly the case when we talk about “shortcut learning”.

“Data is a reflection of the inequalities that exist in the world”. While this might be true, developers have great potential to curb bias in their computer vision systems.

AI (artificial intelligence) is capable of helping the world scale solutions to our biggest challenges but if you haven’t experienced or heard about AI’s mishaps then you’ve been living under a rock. Coded bias, unreliable hospital systems and dangerous robots have littered headlines over the past few years.

Testing machine learning systems is currently more of an art form than a standardized engineering practice. This is particularly problematic for machine learning in mission-critical contexts. This article summarizes three steps from our ML testing series that any development team can take when testing their ML systems.

In this instance of our ML testing series, we discuss fuzz testing. We discuss what it is, how it works, and how it can be used to stress test machine learning systems to gain confidence before going to production.