From the course: Security Risks in AI and Machine Learning: Categorizing Attacks and Failure Modes
Join today to access over 24,700 courses taught by industry experts.
Data bias considerations
From the course: Security Risks in AI and Machine Learning: Categorizing Attacks and Failure Modes
Data bias considerations
- [Instructor] Do you have a favorite movie, one you could watch over and over? If you do, there's a good chance that when you find out someone you're speaking to also loves that movie, that you feel more friendly towards them. Why? Because of bias, a tendency, inclination or preference for or against something. Biases are tricky because they often are based on a feeling or past experience rather than on logical analysis of data. And bias is a difficult subject to discuss because it can be very hard to identify, especially within ourselves. Whether we are aware of it or not, bias exists. And just as personal bias can impact our judgment, biased data can impact the judgment of an ML or AI system leading to unintentional failure. ML systems learn from the data they are trained on. If the training data is biased, the system will automate and repeat that bias. One well known example is discrimination in facial recognition…
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.
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.
