Research
Computational Philosophy
Computational philosophy uses formal models, simulations and data-driven methods to investigate philosophical questions. Across much of my work, these methods help make assumptions explicit, trace their consequences precisely, and explore the dynamics of belief, inquiry, representation and understanding.
Simulated Belousov–Zhabotinsky reaction-diffusion waves generated in Python using a cyclic cellular automaton.
Bayesian Networks, Multi-Belief Networks and Belief Polarisation
Bayesian networks are probabilistic graphical models used to represent complex inferential relationships. My research examines how they can model rational belief polarisation, in which agents' beliefs diverge despite updating on the same evidence, both between pairs of agents and across larger social networks. This work is developed further in my account of rational belief polarisation and factionalisation.
Image credit: Martin Vorel
Foundations of Quantum Field Theory
Haag's theorem appears to challenge standard techniques in quantum field theory, including the interaction picture and perturbation theory. My research, conducted with Marian Gilton and Chris Mitsch, investigates the implications of the theorem for foundational physics and the appropriate methodologies for addressing these challenges. My account of the foundations of quantum field theory places Haag's theorem alongside renormalisation, effective field theory and competing conceptions of foundational work.
Bubble Chamber image from the Lawrence Berkeley National Laboratory
The Epistemology and Ethics of Artificial Intelligence
As AI systems become increasingly sophisticated, understanding how they function and how they should be deployed ethically becomes essential. My research emphasizes that understanding the epistemology of AI is a prerequisite for ethical reasoning about these technologies. My account of artificial intelligence and understanding asks when learned representations support genuine understanding and why that understanding often remains fractured.
Image generated by OpenAI's ChatGPT using DALL·E 3 technology
The Epistemology of Models and Simulations in High Energy Physics
Computational models play crucial roles in high-energy physics research. My work focuses on "sloppy models"—those that depend on numerous parameters but are insensitive to most parameter combinations—and examines their relationship to scientific realism.
CERN Computer Centre (Cern/science Photo Library, 2010)
Reinforcement Learning and Evolution with Invention
I study bargaining games where agents can create novel strategies rather than simply selecting from predetermined options. My research investigates how unsuccessful strategies can diminish through evolutionary and reinforcement learning mechanisms.
Image credit: Jack Moreh
Vector Boson Pair Production at the ATLAS Detector
My prior particle physics work analyzed data from the Large Hadron Collider. This research measured WW and WZ boson pair production cross-sections and investigated anomalous triple gauge couplings. The work involved Monte Carlo simulations and analyses of both merged and separated decay products.
The eight toroid magnets of the ATLAS detector (Maximilien Brice)