TSPLIB with Profits
Python library for parsing TSP with Profits datasets.
Posts by Collection
portfolio
publications
An Expectation-Based Network Scan Statistic for a COVID-19 Early Warning System
Published in ML4PH @ NeurIPS 2020, 2020
Recommended citation: Haycock, C., Thorpe-Woods, E., Walsh, J., O'Hara, P., Giles, O., Dhir, N., and Damoulas, T. (2020). An expectation-based network scan statistic for a covid-19 early warning system. Machine Learning in Public Health workshop, Neural Information Processing Systems. https://arxiv.org/abs/2012.07574
Near Real-Time Social Distance Estimation In London
Published in The Computer Journal, 2024
Recommended citation: James Walsh, Oluwafunmilola Kesa, Andrew Wang, Mihai Ilas, Patrick O’Hara, Oscar Giles, Neil Dhir, Mark Girolami, Theodoros Damoulas, Near Real-Time Social Distance Estimation In London, The Computer Journal, Volume 67, Issue 1, January 2024, Pages 95–109. https://doi.org/10.1093/comjnl/bxac160
talks
Distributionally Robust Routing Problems
Published:
Abstract: Given a probability distribution over the cost function of edges in a graph, a stochastic routing optimisation problem finds a route in the graph that minimises a risk function of the route cost. In practise, it is rare that we have access to the true data-generating distribution over edge costs. Instead, we are given a set of empirical cost observations. From the observations, we can construct a (Bayesian) machine learning model of the costs, then do out-of-sample prediction with our model on new, unseen data. If the observations are noisy or in limited supply - or if our model is misspecified - we might not trust the posterior predictive distribution of the model. In this talk, we discuss routing optimisation algorithms that are robust with respect to an ambiguity set of distributions that are close to the predictive distribution of the model. Real-world applications to air pollution are demonstrated.
teaching
CS259 Formal Languages 20/21
Undergraduate course, Department of Computer Science, University of Warwick, 2020
Seminar tutor for approx 30 students in the 20/21 academic year. Subjects taught include regular languages, context-free languages, and turing machines.
CS259 Formal Languages 21/22
Undergraduate course, Department of Computer Science, University of Warwick, 2021
Seminar tutor for approx 30 students in the 21/22 academic year. Subjects taught include regular languages, context-free languages, and turing machines.