Note: In computer and information science, papers published in major conference proceedings are double-anonymously peer reviewed and recognized as equivalent research contributions to journal articles.

Selected work in progress

• “Path to New Forms of Machine Learning Accountability: Identifying Gap and Challenges in Designing for the Implementation of Social Claim Replicability.” (w/ Dana Calacci)
• “What is the Hype? A Relational Conception: Demonstration Using Four Machine Learning Based Policing Tools.” (w/ Dana Calacci, Nasser Eledroos, and David Gray Widder)

Selected publications

• Kou, Tianqi, Dana Calacci, and Cindy Lin. “Dead Zone of Accountability: Why Social Claims in Machine Learning Research Should Be Articulated and Defended.” Forthcoming in AIES 2025. preprint.
• Kou, Tianqi. “From Model Performance to Claim: How a Change of Focus in Machine Learning Replicability Can Help Bridge the Responsibility Gap.” Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency. 2024. article.

Selected talks

• Tianqi Kou. “Claim Replicability and the Responsibility Gap.” Invited Talk, Digital Life Initiative at Cornell Tech, Roosvelt Island, NYC, 2024.
• Tianqi Kou. “The Function of Replication Studies in Machine Learning Research.” Keynote, Philosophy of Science Meets Machine Learning (PhilML), Tübingen, Germany, 2023.