POSTPONED: CHBH Seminar Series: Dr Max Little

Please note that this event is postponed and will not take place on the above/below date. Please keep an eye out for the rescheduled date!

These seminars are free to attend and are open to all, both within and outside the University. Please register your interest to attend using the link above.

We are delighted to announce that the Centre for Human Brain Health (CHBH) will welcome Dr Max Little, Senior Lecturer in the School of Computer Science, University of Birmingham and Turing Fellow, to present an online CHBH Seminar (via Zoom, with potential for a small physical audience), taking place on Thursday 24th February, 13:00-14:00 GMT.

To arrange a 1:1 meeting or would with Dr Little, please contact us.

If you wish to attend, you can register your interest using the link above.

Causal bootstrapping - plug-in interventional machine learning

To draw scientifically meaningful conclusions and draw reliable inferences of quantitative phenomena, machine learning must take cause and effect into consideration (either implicitly or explicitly). This is particularly challenging when the relevant measurements are not obtained from controlled experimental (interventional) settings, so that cause and effect can be obscured by spurious, indirect influences. Modern predictive techniques from machine learning are capable of capturing high-dimensional, complex nonlinear relationships between variables while relying on few parametric or probabilistic modelling assumptions. However, since these techniques are associational, applied to observational data they are prone to picking up spurious influences from non-experimental (observational) data, making their predictions unreliable. Techniques from causal inference, such as probabilistic causal diagrams and do-calculus, provide powerful (nonparametric) tools for drawing causal inferences from such observational data. However, these techniques are often incompatible with modern, nonparametric machine learning algorithms since they typically require explicit probabilistic models.

In this talk I'll describe causal bootstrapping, a new set of techniques we have developed for augmenting classical nonparametric bootstrap resampling with information about the causal relationship between variables. This makes it possible to resample observational data such that, if it is possible to identify an interventional relationship from that data, new data representing that relationship can be simulated from the original observational data. In this way, we can use modern statistical machine learning and signal processing algorithms unaltered to make statistically powerful, yet causally-robust, inferences.

Speaker Biography

Max Little is a leading expert in statistical machine learning for signal processing and causal inference. His background is in applied mathematics and computer science. Most of his applied work is in biomedical engineering, in particular algorithms for digital health using wearable devices and smartphones. He began his career writing software, signal processing algorithms and music for video games, then moved on by way of a degree in mathematics to the University of Oxford. After postdoc positions in Oxford and co-founding a web-based image search business, he won a Wellcome Trust fellowship at MIT to follow up on his doctoral research work in biomedical signal processing, where he was selected as a TED Fellow. He is currently an Associate Professor at the University of Birmingham and a Fellow of the Alan Turing Institute.

These seminars are free to attend and are open to all, both within and outside the University. Please register your interest to attend using the link above.