Qi Shi 石琪

Research Interest

I am a theoretical researcher with a focus on Agents and Multiagent Systems. I devote my effort to figuring out

• how the agents should act -- AI Ethics, Responsibility, Deontic Logic

• how the agents would act -- Game Theory, Epistemic Game Theory

• how to incentivise/regulate the actions of agents -- Mechanism Design, Normative System

Education

University of Southampton, UK (12.2022 -- now)

• PhD in Computer Science

• Supervisor: Dr Pavel Naumov

• Agents, Interaction and Complexity Research Group

University of Electronic Science and Technology of China, China (09/2019 -- 06/2022)

• Master's Degree in Computer Science and Technology

• Supervisor: Dr Dong Hao

Shanghai University of Finance and Economics, China (09/2015 -- 09/2019)

• Bachelor's Degree in Accounting

•  Second Bachelor's Degree in Computer Science and Technology

Montpellier Business School, France (09/2017 -- 01/2018)

• Undergraduate exchange project, funded by China Scholarship Council

Publications

Journal Articles

1. Qi Shi and Pavel Naumov. Responsibility in multi-step decision schemes. Journal of Philosophical Logic, 54:493-531 2025. 🔗

2. Qi Shi and Dong Hao. Social sourcing: incorporating social networks into crowdsourcing contest design. IEEE/ACM Transactions on Networking, 31:1535–1549, 2023. 🔗

3. Dong Hao, Dongcheng Zhang, Qi Shi, and Kai Li. Entropy regularized actor-critic based multi-agent deep reinforcement learning for stochastic games. Information Sciences, 617:17–40, 2022. 🔗

Conference Proceedings

1. Qi Shi and Pavel Naumov. Uncommon belief in rationality. In Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI-25), 2025.

2. Qi Shi. Responsibility in extensive form games. In Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI-24), volume 38, pages 19920–19928, 2024. 🔗

3. Qi Shi. Agentive permissions in multiagent systems. In Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI-24), pages 3558–3566, 2024. 🔗

4. Qi Shi, Shuai Mao, and Dong Hao. Incentive scoring mechanism design in battle royale games. In Proceedings of the 5th International Conference on Algorithms, Computing and Systems, page 13–19, 2021. 🔗

Preprints

1. Sophia Knight, Pavel Naumov, Qi Shi, and Vigasan Suntharraj. Truth set algebra: a new way to prove undefinability. arXiv:2208.04422, 2022. 🔗

2. Dong Hao, Qi Shi, Jinyan Su, and Bo An. Cooperation, retaliation and forgiveness in revision games. arXiv:2112.02271, 2021. 🔗

Presentations

1. Oral Presentation: Uncommon belief in rationality (03.2025)

at the 39th Annual AAAI Conference on Artificial Intelligence (AAAI-25)

2. Seminar: Uncommon belief in rationality (02.2025)

at Agents, Interaction and Complexity Research Group, University of Southampton

3. Seminar: Responsibility and norms in multiagent systems (09.2024)

at Algorithms and Logic Group, University of Electronic Science and Technology of China

4. Oral Presentation: Agentive permissions in multiagent systems (08.2024)

at the 33rd International Joint Conference on Artificial Intelligence (IJCAI-24)

5. Seminar: Responsibility in extensive form games (02.2023)

at Agents, Interaction and Complexity Research Group, University of Southampton

Teaching

Teaching Assistant

• Automatic software verification (graduate) -- 2023, 2024

marking, slide preparation, coursework preparation

• Algorithmic game theory (graduate) -- 2023, 2024, 2025

marking, tutorial, invigilation

• Algorithmics (undergraduate) -- 2023, 2024, 2025

help (free-question) sessions

• Discreet Maths (undergraduate) -- 2020

marking, remote-class assistance

Teaching

• Programming language - C/C++ (junior) -- 2021

Service

To Community

• PC member/Reviewer for ECAI-24, 25; AAAI-24, 25, 26; IJCAI-25

• Student volunteer for IJCAI-24, AAAI-25

To Institution

• PhD student representative (12.2023 -- 07.2025) 

• Lab Management Assistant (09.2020 --09.2021)