Welcome to the first section of the course. Our goal for this section is to start off with a basic linear model and apply the kernel trick to derive an advanced non-parametric analogue.
We shall cover:
- Projecting data into higher dimensions to obtain linear separability
- Using the kernel trick to derive the Kernel Ridge Regression (KRR) model
- Experimenting with KRR models on different data sets
- Identifying and choosing suitable hyperparameters and deploying the model