About

I am currently a Ph.D. student at Purdue University’s Machine Intelligence and Networked Data Science Group (MINDS) Laboratory advised by Prof. Abolfazl Hashemi. My research focuses on trustworthy AI.

Previously, I have received my M.Sc. degree with a CGPA of 3.91/4.00 from the Department of Electrical and Electronics Engineering, Bilkent University, in 2024, under the supervision of Prof. Suleyman Serdar Kozat. My research focused on modeling and prediction of time series under nonstationary environments. I have received my B.S. degree with high honors and graduated with a CGPA of 3.61/4.00 from the Department of Electrical and Electronics Engineering at Middle East Technical University, Turkey, in 2021.

My research primarily centers on deep learning and large-scale optimization, with an emphasis on developing robust and privacy-preserving deep learning algorithms with mathematical guarantees. Recently, I have been particularly interested in understanding the learning dynamics of large-scale models, such as the memorization phenomenon, and studying how these dynamics evolve under constraints like robustness and adversarial settings. I am also exploring the extension of these theoretical insights to practical domains, including distributed and federated learning.

Research Background & Interests

Research Interests

• Deep Learning (Neural Networks, Sequence Models, Transformers, Generative Networks)
• Optimization (Constrained/Unconstrained Optimization Theory, Multi-Objective Optimization)
• Trustworthy Machine Learning (Robustness, Privacy, Memorization)
• Distributed/Federated Learning

Previous Background

• Online Learning (Transfer Learning, Online Convex Optimization)
• Time Series Analysis and Forecasting (Univariate \& Multivariate Forecasting, State Space Models, Anomaly Detection)