Machine Learning

Machine learning is the study of computational processes that find patterns and structure in data.

Our group is interested in a broad range of theoretical aspects of machine learning as well as applications. Much of the current excitement around machine learning is due to its impact in a broad range of applications. The applications considered in our research include astronomy, systems biology, neuroscience, natural language processing, robotics, and computer vision.

Faculty

Members	Research interests
Amos Storkey	Continuous time systems, deep learning, stochastic differential equations
Iain Murray	Bayesian statistics, approximate inference, Markov chain Monte Carlo, scientific data analysis
Siddharth Narayanaswamy	Explainable and Interpretable ML, Bayesian program learning, Vision and Language, Probabilistic programming, Neuro-symbolic systems, Approximate inference, Human-machine interaction.
Nigel Goddard	Probabilistic modeling of energy-related systems
Chris Bishop	Graphical models, variational methods, pattern recognition
Michael Gutmann	Efficient statistical learning, inference for complex models, unsupervised deep learning, natural image statistics, computational biology
Arno Onken	Probabilistic models, in particular copula-based models; Dimensionality reduction techniques; Information theory; Applications to biological systems
Ava Khamseh	Semi-parametric probabilistic modelling, targeted learning, causal inference and its applications to population biomedicine and cancer research
Oisin Mac Aodha	Human-in-the-loop machine learning, machine teaching, deep learning, and computer vision
Nikolay Malkin	Bayesian machine learning and generative modelling, amortised inference for (neuro-)symbolic models, probabilistic reasoning/planning in language and formal systems, AI for science and mathematics.
Antonio Vergari	Efficient and reliable machine learning in the wild, tractable probabilistic modeling, combining learning and reasoning
Chris Williams	Gaussian processes, image interpretation, unsupervised learning, deep learning, time series models