From the course: AI Accountability: Build Responsible and Transparent Systems (2022)
Unlock the full course today
Join today to access over 24,800 courses taught by industry experts.
The causes of classification errors
From the course: AI Accountability: Build Responsible and Transparent Systems (2022)
The causes of classification errors
- [Instructor] There's a saying in the book publishing world that for each equation you include in your book, you're going to lose 50% of your audience. I have to talk about a mathematical topic but I'm going to try to do it in a way that doesn't rely on equations because I'm hoping that you will all stick around. Let me introduce you to Thomas Bayes, a Presbyterian minister as well as philosopher and statistician who lived in England in the 1700s. and Bayes is best known for Bayes' Theorem, a particular formula he created that gives you the probability of an accurate conclusion given the data. What's amazing that this is not normally how statistics works, this is a variation on it but it requires that you know a few things in order to calculate this probability. Number one is you have to know the prevalence or the base rate of the phenomenon. Think about diagnosing the disease. You have to know what percentage of…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
The challenge of classification errors3m 39s
-
(Locked)
The causes of classification errors6m 18s
-
(Locked)
Bias in AI3m 50s
-
(Locked)
Supervised and unsupervised learning8m 16s
-
(Locked)
Biased labeling of data7m 15s
-
(Locked)
Construct validity6m 14s
-
(Locked)
The absence of meaning4m 54s
-
(Locked)
Vulnerability to attacks4m 45s
-
-
-
-
-
-