Skip to 0 minutes and 1 second OK. And welcome back to week two of the Advanced Machine Learning course here on “Future Learn.” So this week, we’re going to move on from background theory that we looked at in week one, and we’re going to look at how you can evaluate models, which models are the best of the set that you create, which model should you select to use, and evaluating the expected performance of that selected model on your data. We’ll also look at how to evaluate the data that you have. Do you have enough data? Do you have enough training data?
Skip to 0 minutes and 31 seconds Do you have enough validation data to be able to evaluate which model is better than others to be statistically confident that the model you select is genuinely better than the second best model, for example? And how to work out if you’ve got enough test data to get a good and accurate estimate of the performance of your final model. After those topics, we’ll move on to actually looking at modelling algorithms. In particular, we’re going to be looking at supervised learning algorithms. This week, we’ll be looking at artificial neural networks. We’re going to look at them in depth. We’ll see a lot about them, what they are, how they work, how they’re trained, how they’re regularised.
Skip to 1 minute and 14 seconds There will be some example exercises that we’ll go through, and there will be a video looking at how these ideas that we’re looking at this course can be extended and applied when we look at a deep learning toolkit. OK. So I hope you’re going to enjoy this week, and let’s get to it.
Introduction to Week 2
A short introduction for week two of the course.
© Dr Michael Ashcroft