Professor David LeslieProfessor of Statistics
I am a Professor of Statistical Learning in the Department of Mathematics and Statistics at Lancaster University. I research statistical learning, decision-making, and game theory. My research on bandit algorithms is used by many of the world's largest companies to balance exploration and exploitation in real time website optimisation. I collaborate with several companies, including Prowler.io and BT (the latter through the NG-CDI project, funded by an EPSRC Prosperity Partnership). I also lead the EPSRC/NERC-funded Data Science of the Natural Environment (DSNE) project at Lancaster University. Prior to my position at Lancaster, I was a senior lecturer in the statistics group of the School of Mathematics, University of Bristol, where I was co-director of the EPSRC-funded cross-disciplinary decision-making research group at the University of Bristol. I was also a partner in the ALADDIN project, a large strategic partnership between BAE Systems and EPSRC, and involving researchers from Imperial College, Southampton, Oxford, Bristol and BAE Systems.
- statistical learning
- game theory, particularly learning in games
- reinforcement learning
- stochastic approximation
I have held numerous roles at the Royal Statistical Society. These include elected Council member, member of the long term strategy group, member of the Academic Affairs Advisory Group, and chair of the Applied Probability Section, and member of the Research Section Committee.
I have served on the programme committe of the London Mathematical Society.
PhD Supervisions Completed
I have had the privilege to supervise some great students. The following have finished and moved on:
- Archie Chapman
- Michalis Smyrnakis
- Harriet Mills
- Ben May
- Steve Perkins
- Kevin Lloyd
- Matthew Arnold
- Simon Smith
- Adnane Ez-zizi
- Beki Floyd
- Hannah Julienne
- James Edwards
- Chris Sherfield
I am always on the lookout for the next good research student. However, note that it is essentially impossible for me to find financial support for students from outside the EU.
A small amount of further information is available on my personal webpage
Bandit learning in concave N-player games
Bravo, M., Leslie, D.S., Mertikopoulos, P. 2/12/2018
Selecting Multiple Web Adverts - a Contextual Multi-armed Bandit with State Uncertainty
Leslie, D.S., Edwards, J.A. 2/01/2020 In: Journal of the Operational Research Society. 71, 1, p. 100-116. 17 p.
Using J-K-fold Cross Validation to Reduce Variance When Tuning NLP Models
Moss, H., Leslie, D.S., Rayson, P.E. 06/2018
Robustness Properties in Fictitious-Play-Type Algorithms
Swenson, B., Kar, S., Xavier, J., Leslie, D.S. 24/10/2017 In: SIAM Journal on Control and Optimization. 55, p. 3295-3318. 24 p.
Mixed-strategy learning with continuous action sets
Perkins, S., Mertikopoulos, P., Leslie, D.S. 01/2017 In: IEEE Transactions on Automatic Control. 62, 1, p. 379-384. 6 p.
REX: a development platform and online learning approach for Runtime emergent software systems
Porter, B.F., Grieves, M., Rodrigues Filho, R., Leslie, D.S. 2/11/2016
Optimistic Bayesian sampling in contextual-bandit problems
May, B.C., Korda, N., Lee, A., Leslie, D.S. 06/2012 In: Journal of Machine Learning Research. 13, p. 2069-2106. 37 p.
DSI: Research supervision for Prowler.io - contract renewal
01/03/2019 → 29/02/2020
DSI: Data Science of the Natural Environment
16/04/2018 → 15/04/2024
DSI : Contextual Bandits for Retail Pricing
01/03/2017 → 31/12/2018
STOR-i Centre for Doctoral Training
- Analysis and Probability
- DSI - Foundations
- Statistical Learning
- STOR-i Centre for Doctoral Training