From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Join today to access over 23,200 courses taught by industry experts.
Supervised learning vs. unsupervised learning
From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Supervised learning vs. unsupervised learning
- [Instructor] Now that we know what artificial intelligence is and what we need to create AI or ML models let's talk about the big distinction in the type of ML methods that are supervised and unsupervised learning. In supervised methods, we have a mechanism of telling the model what the target is. This could be a numeric value or whether something belongs to a specific class or if something is correct or incorrect. We call these targets, labels. These labels are a specific target for the model to predict and we can evaluate the model on how accurately it predicted the labels. If, for instance, we know that a picture is going to be one of the three different categories, a horse, a dog, or a cat we can label those images as such and then test our model out against that pre-labeled data. Because we have decided that these labels are correct we can use them to get a precise notion of what error the model has. One of the trade-offs…
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.