Jiarui Gan

Jiarui Gan
Wolfson Building, Parks Road, Oxford OX1 3QD
Interests
I work in computational game theory, multi-agent systems, and AI. My research is motivated by quests for good strategies for interacting with intelligent agents, especially in complex real-world settings. I am fascinated by the power of computation to address these challenges and enjoy solving the algorithmic “puzzles” that emerge along the way. By analysing the fundamental computational properties of these puzzles, my work uncovers what is and is not computationally feasible. We develop algorithmic solutions and theories for steering/guiding/coordinating agents toward desirable outcomes. The broader vision is to ensure that AI agents in next-generation agentic systems are not only more powerful, but also more principled, cooperative, and better aligned with the common good.
🔍 I'm looking for PhD students. Feel free to get in touch if you're interested! (See more about PhD at Oxford here.) For any application-related inquiries, please directly contact me (jiarui.gan@cs.ox.ac.uk) or the Department of Computer Science.
Biography
Prior to rejoining Oxford, I was a postdoctoral researcher at Max Planck Institute for Software Systems in Germany. I obtained my DPhil degree at Oxford in 2021.
Selected Publications
-
Bayesian persuasion in sequential decision−making
Jiarui Gan‚ Rupak Majumdar‚ Goran Radanovic and Adish Singla
In Proceedings of the 36th AAAI Conference on Artificial Intelligence (AAAI'22). Pages 5025–5033. 2022.
Details about Bayesian persuasion in sequential decision−making | BibTeX data for Bayesian persuasion in sequential decision−making
-
Schelling games on graphs
Aishwarya Agarwal‚ Edith Elkind‚ Jiarui Gan‚ Ayumi Igarashi‚ Warut Suksompong and Alexandros A Voudouris
In Artificial Intelligence. Vol. 301. Pages 103576. 2021.
Details about Schelling games on graphs | BibTeX data for Schelling games on graphs
-
Optimally deceiving a learning leader in stackelberg games
Georgios Birmpas‚ Jiarui Gan‚ Alexandros Hollender‚ Francisco Marmolejo‚ Ninad Rajgopal and Alexandros Voudouris
In Advances in Neural Information Processing Systems (NeurIPS'20). Pages 20624–20635. 2020.
Details about Optimally deceiving a learning leader in stackelberg games | BibTeX data for Optimally deceiving a learning leader in stackelberg games