Personal Website - Arif Kerem Dayı

About me

About me

I am a first year Ph.D student at MIT CSAIL, working with Profs Constantinos Daskalakis and Russ Tedrake. My research interests lie in the theory of diffusion models, and its applications to improved sampling in sequence models in video generation, robotics, etc.

Previously, I received an A.B. in Computer Science and Mathematics, and S.M in Computer Science from Harvard University in May 2025. During my undergraduate education, I was fortunate to work on the theory of low-rank fine tuning with SGD (advised by Prof. Sitan Chen), and various problems in distributed/multi-agent optimization (advised by Profs Stephanie Gil and Angelia Nedich).

Publications

Below, you can find a list of selected publications. For a full list, please visit my Google Scholar page.

Learning trust over directed graphs in multiagent systems

Conference: Learning for Dynamics and Control Conference (L4DC) - 2023

Authors: Orhan Eren Akgun, Arif Kerem Dayı, Stephanie Gil, Angelia Nedich

Low-rank fine-tuning lies between lazy training and feature learning

Conference: Conference on Learning Theory (COLT) - 2025

Authors: Arif Kerem Dayı, Sitan Chen

Projected push-pull for distributed constrained optimization over time-varying directed graphs

Conference: American Control Conference (ACC) - 2024

Authors: Orhan Eren Akgun*, Arif Kerem Dayı*, Stephanie Gil, Angelia Nedich

Teaching

During my time at Harvard, I got the opportunity to serve the CS and math community as a course assistant. I am grateful to have had the chance to help teach the following Harvard classes:

CS 121: Introduction to Theoretical Computer Science: Introductory course covering topics in theory of computation including circuits, turing machines, computability, complexity, reductions and randomized computations, taught by Boaz Barak.

Math 25b: Introduction to Theoretical Linear Algebra and Real Analysis II: A rigorous introduction to single and multi variable real analysis, metric space topology, and basic fourier analysis, taught by Wes Cain.

Math 25a Introduction to Theoretical Linear Algebra and Real Analysis II: A rigorous introduction to linear algebra based on the amazing book Linear Algebra Done Right, taught by Wes Cain.

ES 150: Introduction to Probability with Engineering Applications: Introduction to probability class covering fundamental probability concepts and their applications to engineering problems, taught by Yue M. Lu.