Machine Learning Tutorials

In the Machine Learning Tutorial Series, external guest speakers will give tutorial lectures on focused machine learning topics. The target audience are undergraduates, MSc and PhD students, post-docs and interested faculty members.
All talks will be announced via the ml-talks mailing list.

If you are looking for previous tutorials, check out the ML Tutorials Archive.

Autumn 2018

Normally, the talks will be on Wednesdays, 14:00 – 16:00.



 Date Speaker Title
2018-11-28 Dino Sejdinovic (University of Oxford) Approximate Kernel Methods, Embeddings, and Aggregates
2018-11-07 Tor Lattimore (DeepMind) Bandit algorithms and Reinforcement learning



Approximate Kernel Methods, Embeddings, and Aggregates (Dino Sejdinovic, 2018-11-28 )
Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting probability metric, are useful tools for fully nonparametric hypothesis testing, for the two-sample problem, (conditional) independence testing, and for multivariate interaction. Moreover, they also found applications in learning and predicting on distributional inputs corresponding to the situation where the access to outputs is at a much coarser (aggregate) level. I will give an overview of this framework and give some recent developments in the context of large-scale kernel approximations and weakly supervised learning on aggregates.

Bandit algorithms and Reinforcement learning (Tor Lattimore, 2018-11-07)

Decision making in the face of uncertainty is a significant challenge in machine learning. Which drugs should a patient receive? How should I allocate my study time between courses? Which version of a website will return the most revenue? What move should be considered next when playing chess/go? All of these questions can be expressed in the multi-armed bandit framework where a learning agent sequentially takes actions, observes rewards and aims to maximise the total reward over a period of time. The framework is now very popular, used in practice by big companies, and growing fast. The focus of the tutorial will be on understanding the statistical ideas, mathematics and implementation details of the core algorithmic concepts.