3 Lectures (4.5 hours) given at the 2018 Machine Learning Summer School, Madrid.
Probabilistic machine learning approaches task of describing of data, to complex systems or our world using the language and tools of probability. Almost all of machine learning can be viewed in probabilistic terms, making probabilistic thinking fundamental. It is, of course, not the only view. But it is through this view that we can connect what we do in machine learning to every other computational science, whether that be in stochastic optimisation, control theory, operations research, econometrics, information theory, statistical physics or bio-statistics. For this reason alone, mastery of probabilistic thinking is essential.
The aim of this tutorial is to develop flexible and broad tools that will support your probabilistic thinking. Part 1, Foundations looks at the philosophy of machine learning, builds an understanding of the model-inference-algorithm paradigm, and the explores fundamental areas of machine learning – we’ll look at deep learning, kernels and reinforcement learning. Part 2 Tricks, will look at 6 individual probabilistic problems and a tricks to solve them, using these tricks to develop flexibility in our thinking. Part 3: Algorithms will look at how the foundations and tricks combine to develop machine learning algorithms, with a specific focus on the area of deep generative models.