I work on probabilistic machine learning, recently focusing on normalizing flows and uncertainty in deep learning.
Ordering Dimensions with Nested Dropout Normalizing Flows
Artur Bekasov, Iain Murray
Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models, ICML, 2020 (spotlight)
Research interests / bio
I am interested in probabilistic approaches to machine learning, especially the interplay between deep learning and Bayesian inference. My other interests include unsupervised representation learning, information theory of machine learning, and modelling invariances/symmetries for data-efficient learning.
I am a part of the Centre for Doctoral Training in Data Science, a combined MSc by Research + PhD programme. I have worked on generative video modelling for my MSc thesis.
Before starting on the programme, I worked on recommendations at Amazon and studied Computer Science at the University of Manchester. I have worked on evolutionary computation for my final-year project.
I have tutored and marked for the following courses:
- Machine Learning and Pattern Recognition (2017, 2018, 2019)
- Probabilistic Modelling and Reasoning (2018)
- Computer Programming for Speech and Language (2017, 2018)