Uploaded March 2018 | Updated November 2026, 5 days ago
In this lecture, we first review some of the more important details of the linear model, building up our intuition for what it does, and how it works. We then tackle support vector machines in two ways: first as a loss function for general models, and second as a separate classifier, that allows us to apply something called the kernel trick. Lecturer: Peter Bloem. See the PDF for image credits.
Corrections:
- In slide 72, there is a factor of sqrt(2) missing in the last extended feature.
Slides: dropbox.com/s/fo1vq0hcpne9yq4/41.LinearModels2.annotated.pdf?dl=0
In this lecture, we first review some of the more important details of the linear model, building up our intuition for what it does, and how it works. We then tackle support vector machines in two ways: first as a loss function for general models, and second as a separate classifier, that allows us to apply something called the kernel trick. Lecturer: Peter Bloem. See the PDF for image credits.
Corrections:
- In slide 72, there is a factor of sqrt(2) missing in the last extended feature.
Slides: dropbox.com/s/fo1vq0hcpne9yq4/41.LinearModels2.annotated.pdf?dl=0










