AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How to Monitor Your Machine Learning System for Future Failures
Google recently released a framework of machine learning test scores, which looks at the minimum score you could have gotten along either your feature monitoring feature health about your model monitoring and so on. That would be a first start to even figuring out with respect to if I made the health of something that is being deployed in production. And then there is all the choices that our models make, right? Like because we were going after a particular product objective. Now, what is the longer term impact that this may have overall in our user base? Is it going to be unfair to some segments of our population? "That's an even bigger ethical fairness question," he says.