Hello! I’m Dimitris (feel free to call me Dimitri), a final-year PhD student in Computer Science at Johns Hopkins University (JHU), advised by Prof. Nicolas Loizou. I expect to graduate in Spring 2027 and I am on the job market for research scientist positions starting in 2027.

I’m broadly interested in large-scale optimization and machine learning. My current research focuses on adaptive step-size methods and sharpness-aware minimization, with applications to deep learning and large language model training.

Before JHU, I earned a BSc in Mathematics from the National and Kapodistrian University of Athens (NKUA), and MSc degrees in Data Science and Machine Learning from the National Technical University of Athens (NTUA; thesis) and in Mathematics from the University of Göttingen (thesis).

CV / Email

Publications

Curvature-Scaled Sharpness-Aware Minimization: Convergence Beyond Smoothness
Under Submission
Dimitris Oikonomou, Nicolas Loizou

Taking the Road Less Scheduled with Adaptive Polyak Steps
In the 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026)
Dimitris Oikonomou, Matthew Buchholz, Yuen-Man Pun, Robert M. Gower, Nicolas Loizou

Adaptive Sharpness-Aware Minimization with a Polyak-type Step size: A Theory-Grounded Scheduler
In the 43rd International Conference on Machine Learning (ICML 2026), Code
Dimitris Oikonomou, Nicolas Loizou

Safeguarded Stochastic Polyak Step Sizes for Non-smooth Optimization: Robust Performance Without Small (Sub)Gradients
In the 43rd International Conference on Machine Learning (ICML 2026), Code
Dimitris Oikonomou, Nicolas Loizou

Analysis of an Idealized Stochastic Polyak Method and its Application to Black-Box Model Distillation
In Transactions on Machine Learning Research, 2026 (TMLR)
Robert M. Gower, Guillaume Garrigos, Nicolas Loizou, Dimitris Oikonomou, Konstantin Mishchenko, Fabian Schaipp

Sharpness-Aware Minimization: General Analysis and Improved Rates
In the 13th International Conference on Learning Representations (ICLR 2025), Code
Dimitris Oikonomou, Nicolas Loizou

Stochastic Polyak Step-sizes and Momentum: Convergence Guarantees and Practical Performance
In the 13th International Conference on Learning Representations (ICLR 2025), Code
Dimitris Oikonomou, Nicolas Loizou

Honors & Awards

Mathematical Olympiads

• International Mathematical Olympiad (IMO): Bronze Medal @ Santa Marta, Colombia, 2013
• Greek Mathematical Olympiad: Silver Medal @ Athens, Greece, 2014
• Greek Mathematical Olympiad: Bronze Medal @ Athens, Greece, 2013

Awards & Scholarships

• Gold Reviewer Award, International Conference on Machine Learning (ICML 2026)
• JHU MINDS Data Science Fellowship: 2024, 2025
• DAAD Scholarship (German Academic Exchange Service): 2019-2020

Talks & Presentations

• Poster, NeurIPS 2026, Atlanta, USA, December 2026 (upcoming)
• 2× Poster, ICML 2026, Seoul, South Korea, July 2026
• Invited Talk, SIAM Conference on Optimization (OP26), Edinburgh, UK, June 2026: Bridging Momentum and Polyak Steps: Recent Theoretical and Practical Advances
• Invited Talk, ELLIIT Focus Period on Optimization for Learning, Lund University, Sweden, May 2026
• Poster, ELLIIT Symposium on Optimization for Learning, Lund University, Sweden, May 2026
• 2× Poster, Greeks in AI 2025, Athens, Greece, July 2025
• 2× Poster, ICLR 2025, Singapore, April 2025

Teaching & Professional Service

• Head Teaching Assistant: Computer Vision @ JHU, Fall 2024
• Mentor: JHU MINDS summer outreach program (high-school student), Summer 2024
• Reviewer: ICML, NeurIPS, ICLR, AISTATS; TMLR, IEEE Transactions on Signal Processing, Information and Inference

Last Update: October 7, 2026