From the course: Threat Modeling: Repudiation in Depth
Join today to access over 22,500 courses taught by industry experts or purchase this course individually.
AI and machine learning
From the course: Threat Modeling: Repudiation in Depth
AI and machine learning
- [Instructor] There're a lot of deep learning systems whose results are hard to understand. And it's okay if sometimes your pictures of cats also include a picture of a sweater or a teacup. But when your AI is making more impactful decisions, and I hope it is, then some of those decisions are going to be ones that people want to understand, contest or reject. So for example, if you're a payment processor, using an AI to make decisions about what transactions to refuse, your merchants are going to want details. If you're delivering spam filters, the people whose emails are blocked are going to want to know why. Sometimes, this is referred to as explainable AI or transparency. It's not always cheap or easy but that doesn't mean it's not important. Also important, you want to test your AI for viruses including those that come from the training data. As I was prepping this course, there was a story in the news about…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.