Sarah L. Thomson
sarah l thomson

Dr Sarah L. Thomson

Lecturer

  S.Thomson4@napier.ac.uk

  

Biography

Dr Sarah L. Thomson is a lecturer in data science at Edinburgh Napier University, having started there in August 2023. She was previously at the University of Stirling: she was awarded her PhD there in March 2021, and worked as a research fellow from September 2020 until May 2022; after that. she took up a lectureship in June 2022, before moving to Edinburgh Napier in August 2023.

Dr Thomson's expertise is predominantly in evolutionary computation, optimisation, and machine learning. She has worked on problems from healthcare, aviation, vehicle management, and agriculture. Additionally, she has a passion for fundamental research and is particularly interested in fitness landscapes and explainable artificial intelligence (XAI).

School

Research Groups

Research Areas

Filter by

Date


28 results

Subfunction Structure Matters: A New Perspective on Local Optima Networks

Presentation / Conference Contribution
Thomson, S. L., & Przewozniczek, M. W. (2025, July)
Subfunction Structure Matters: A New Perspective on Local Optima Networks. Presented at Genetic and Evolutionary Computation Conference (GECCO 2025), Málaga, Spain
Local optima networks (LONs) capture fitness landscape information. They are typically constructed in a black-box manner; information about the problem structure is not utilis...

Into the Black Box: Mining Variable Importance with XAI

Presentation / Conference Contribution
Hunter, K., Thomson, S. L., & Hart, E. (2025, April)
Into the Black Box: Mining Variable Importance with XAI. Presented at Evostar 2025, Trieste, Italy
Recent works have shown that the idea of mining search spaces to train machine learning models can facilitate increasing understanding of variable importance in optimisation p...

Stalling in Space: Attractor Analysis for any Algorithm

Presentation / Conference Contribution
Thomson, S. L., Renau, Q., Vermetten, D., Hart, E., van Stein, N., & Kononova, A. V. (2025, April)
Stalling in Space: Attractor Analysis for any Algorithm. Paper presented at EvoStar 2025, Trieste, Italy
Network-based representations of fitness landscapes have grown in popularity in the past decade; this is probably because of growing interest in explainability for optimisatio...

A Deep Dive into Effects of Structural Bias on CMA-ES Performance along Affine Trajectories

Presentation / Conference Contribution
van Stein, N., Thomson, S. L., & Kononova, A. V. (2024, September)
A Deep Dive into Effects of Structural Bias on CMA-ES Performance along Affine Trajectories. Presented at Parallel Problem Solving from Nature (PPSN) 2024, Hagenberg, Austria
To guide the design of better iterative optimisation heuristics, it is imperative to understand how inherent structural biases within algorithm components affect the performan...

Entropy, Search Trajectories, and Explainability for Frequency Fitness Assignment

Presentation / Conference Contribution
Thomson, S. L., Ochoa, G., van den Berg, D., Liang, T., & Weise, T. (2024, September)
Entropy, Search Trajectories, and Explainability for Frequency Fitness Assignment. Presented at Parallel Problem Solving from Nature (PPSN 2024), Hagenberg, Austria
Local optima are a menace that can trap optimisation processes. Frequency fitness assignment (FFA) is an concept aiming to overcome this problem. It steers the search towards ...

Addressing the traveling salesperson problem with frequency fitness assignment and hybrid algorithms

Journal Article
Liang, T., Wu, Z., Lässig, J., van den Berg, D., Thomson, S. L., & Weise, T. (2024)
Addressing the traveling salesperson problem with frequency fitness assignment and hybrid algorithms. Soft Computing, 28, 9495–9508. https://doi.org/10.1007/s00500-024-09718-8
The traveling salesperson problem (TSP) is one of the most iconic hard optimization tasks. With frequency fitness assignment (FFA), a new approach to optimization has recently...

Factors Impacting Landscape Ruggedness in Control Problems: a Case Study

Presentation / Conference Contribution
Saliby, M. E., Medvet, E., Nadizar, G., Salvato, E., & Thomson, S. L. (2024, September)
Factors Impacting Landscape Ruggedness in Control Problems: a Case Study. Presented at WIVACE 2024 (XVIII International Workshop on Artificial Life and Evolutionary Computation), Namur, Belgium
Understanding fitness landscapes in evolutionary robotics (ER) can provide valuable insights into the considered robotic problems as well as into the strategies found by evolu...

Where the Really Hard Quadratic Assignment Problems Are: the QAP-SAT instances

Presentation / Conference Contribution
Verel, S., Thomson, S. L., & Rifki, O. (2024, April)
Where the Really Hard Quadratic Assignment Problems Are: the QAP-SAT instances. Presented at EvoCOP 2024, Aberystwyth, UK
The Quadratic Assignment Problem (QAP) is one of the major domains in the field of evolutionary computation, and more widely in combinatorial optimization. This paper studies ...

Shape of the Waterfall: Solvability Transitions in the QAP

Presentation / Conference Contribution
Akova, S., Thomson, S. L., Verel, S., Rifki, O., & van den Berg, D. (2024, April)
Shape of the Waterfall: Solvability Transitions in the QAP. Presented at EvoStar 2024, Aberyswyth, Wales
We consider a special formulation of the quadratic assignment problem (QAP): QAP-SAT, where the QAP is composed of smaller sub-problems or clauses which can be satisfied. A re...

Frequency Fitness Assignment for Untangling Proteins in 2D

Presentation / Conference Contribution
Koutstaal, J., Kommandeur, J., Timmer, R., Horn, R., Thomson, S. L., & van den Berg, D. (2024, April)
Frequency Fitness Assignment for Untangling Proteins in 2D. Presented at EvoStar 2024, Aberyswyth, UK
At the time of writing, there is no known deterministic-time algorithm to sample valid initial solutions with uniform random distribution for the HP protein folding model, bec...

Current Post Grad projects