Search

Machine learning

Learn more

Projects

Capturing complex data streams

Describing complex sequences of data from different sources, to gain insights and generate meaningful actions in a range of applications

Cancer pre-diagnostic analytics with AI

Creating a digital repository of a variety of tumour and immune cells and developing algorithms to recognise these cells automatically

Bayesian analysis of single cell sequencing data

Developing statistical tools for biological data, to understand complex biological processes, such as cancer

Artificial intelligence for data analytics (AIDA)

Drawing on new advances in AI and machine learning to address data wrangling issues, and help to automate the data analytics process

Anomaly detection to combat fraud

Finding patterns in data that depart significantly from expected behaviour, in order to make fraud detection quicker and more cost-efficient

Adversarial machine learning

Studying how systems trained with machine learning can be affected by manipulated data, such as faked biometrics and hidden malware

Publications and software

Research publications

Modeling outcomes of soccer matches

We compare various extensions of the Bradley–Terry model and a hierarchical Poisson log-linear model...

Tsokos, A., Narayanan, S., Kosmidis, I. et al. Mach Learn (2019) 108: 77. https://doi.org/10.1007/s10994-018-5741-1

Research publications

Discovering Interpretable Representations for Both Deep Generative and Discriminative Models

Interpretability of representations in both deep generative and discriminative models is highly desirable. Current...

Tameem Adel, Zoubin Ghahramani, Adrian Weller; Proceedings of the 35th International Conference on Machine Learning, PMLR 80:50-59, 2018.

Research publications

Blind Justice: Fairness with Encrypted Sensitive Attributes

Recent work has explored how to train machine learning models which do not discriminate...

Kilbertus, N., Gascon, A., Kusner, M., Veale, M., Gummadi, K.P. & Weller, A.. (2018). Blind Justice: Fairness with Encrypted Sensitive Attributes. Proceedings of the 35th International Conference on Machine Learning, in PMLR 80:2635-2644

Research publications

TAPAS: Tricks to Accelerate (encrypted) Prediction As a Service

Machine learning methods are widely used for a variety of prediction problems. Prediction as...

Sanyal, A., Kusner, M.J., Gascón, A., & Kanade, V. (2018). TAPAS: Tricks to Accelerate (encrypted) Prediction As a Service. CoRR, abs/1806.03461.

Research publications

Handling Incomplete Heterogeneous Data using VAEs

Variational autoencoders (VAEs), as well as other generative models, have been shown to be...

A. Nazábal, P. M. Olmos, Z. Ghahramani, I. Valera "Handling Incomplete Heterogeneous Data using VAEs", Arxiv preprint, July 2018.

Research publications

A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices

Discrimination via algorithmic decision making has received considerable attention. Prior work largely focuses on...

A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices Till Speicher, Hoda Heidari, Nina Grgic-Hlaca, Krishna P. Gummadi, Adish Singla, Adrian Weller, Muhammad Bilal Zafar KDD '18: The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining 2018, London, United Kingdom, August 2018

Research publications

Labelling as an unsupervised learning problem

Unravelling hidden patterns in datasets is a classical problem with many potential applications. In...

Lyons, Terry, and Imanol Perez Arribas. "Labelling as an unsupervised learning problem." arXiv preprint arXiv:1805.03911 (2018).

Research publications

Sparse communication for distributed gradient descent

We make distributed stochastic gradient descent faster by exchanging sparse updates instead of dense...

Aji, A & Heafield, K 2017, Sparse Communication for Distributed Gradient Descent. in Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (EMNLP 2017). Association for Computational Linguistics (ACL), pp. 440-445, EMNLP 2017: Conference on Empirical Methods in Natural Language Processing, Copenhagen, Denmark, 7-11 September.

Research publications

Beyond Distributive Fairness in Algorithmic Decision Making: Feature Selection for Procedurally Fair Learning

With widespread use of machine learning methods in numerous domains involving humans, several studies...

Grgic-Hlaca, N., Zafar, M.B., Gummadi, K.P., & Weller, A. (2018). Beyond Distributive Fairness in Algorithmic Decision Making: Feature Selection for Procedurally Fair Learning. AAAI.