Research & Projects
Research & Projects
My interests lie in the intersections between game theory, reinforcement learning and microeconomic theory. I analyse environments where many reinforcement learning agents interact, and I look for the four C's: competition, collusion, convergence and chaos.
In Progress (drafts available upon request):
September 2022: Reinforcement Learning in Evolutionary Pricing Games
With Bernhard von Stengel, Galit Ashkenazi-Golan, Rahul Savani, Katerina Papadaki and Sahar Jahani. Presented at ALGA 2023 and the Alpine Game Theory Symposium.
Theoretical and empirical results from applying deep reinforcement learning to a multi-round pricing duopoly model.
Completed:
November 2024: Reinforcement Learning, Collusion and the Folk Theorem
With Galit Ashkenazi-Golan and Domenico Mergoni. Presented at the LSE Seminar on Combinatorics, Games and Optimisation, ALGA 2023, SAET 2023 and the Alpine Game Theory Symposium.
Proves that agents using a class of policy-based reinforcement learning methods can learn feasible and individually rational Nash equilibria in repeated games, even under uncertainty and imperfect monitoring. Link.
July 2023: Analysis of Darcy's Law
With Torin Fastnedge, Yaqing Luo, Jimmy Chen, Nora Alshammari and Atrayee Bhattacharya. Presented at the UK Graduate Modelling Camp 2023. I was awarded the 'Institute of Mathematics Top Student Prize' for this work.
Provides conditions for when voids form in porous media. Link.
May 2023: Reinforcement Learning in Auctions
With Leonard Hinckeldey.
Provides empirical results for the application of Q-Learning and policy gradient methods to auctions. Link.
This is a promising research direction. Please reach out if you are interested.
June 2022: Dynamic Bayesian Persuasion
An analysis of a partially observable dynamic Bayesian persuasion model. Link.
For this work, I was award the 'Haya Freedman Prize'.
July 2021: Equilibria in the Prisoner's Dilemma under Limited Recall
Written under the supervision of Costas Cavounidis.
An analysis of how perfect recall affects equilibrium strategies in the repeated prisoner's dilemma. Link.
April 2021: Computational Approach to Evolutionary Game Theory
With Jacobus Smit.
An analysis of evolutionary stable strategies using agent based modelling and analytical methods. Link.
Conferences and Workshops:
Mediterranean Symposium on Game Theory. 28th -31st May 2024. L’Ecole d’Economie d’Aix-Marseille, Marseille, France.
Stochastic Games: Theoretical and Computational Aspects. 26-28th May 2024. Centre National de la Recherche Scientifique, Fréjus, France.
PhD Research Festival. 24th May 2024. London School of Economics, London, UK. Best Presentation Prize.
What is Strategic Information? A Workshop on Game Theory. 15-16th April 2024. London School of Economics, London, UK.
Alan Turing Institute Enrichment Scheme Day. 24th November 2023. The Alan Turing Institute, London, UK.
Workshop on Game Theory and Machine Learning. 19-20th October 2023. London School of Economics, London, UK.
UK Graduate Modelling Camp. 10-14th July 2023. Robinson College, Cambridge, UK.
Alpine Game Theory Symposium. 26-30th June 2023. Université Grenoble Alpes, Grenoble, France.
Algorithms, Learning, and Games (ALGA) 2023. 5-9th June 2023. Sampieri, Sicily, Italy.
LSE Winter Workshop in Combinatorics. 12-16th December 2022. London School of Economics, London, UK.
What is The Game? A Workshop on Game Theory. 5-6th April 2022. London School of Economics, London, UK.