Arash Jamshidi

I am an ELLIS Ph.D. student in Computer Science at the University of Helsinki, with an exchange host at KTH Royal Institute of Technology. I have a master's in Artificial Intelligence with a focus on Theoretical Machine Learning from Sharif University of Technology. In my Ph.D., I am supervised by Kai Puolamäki and Aristides Gionis, and in my master's I was supervised by Mahdi Jafari Siavoshani. My research focuses on Machine Learning Theory and Trustworthy Machine Learning.

CV | LinkedIn | Email

Research Interests

machine learning theory, high-dimensional statistics, submodularity, robustness in machine learning and statistics.

Publications and Manuscripts

[1] Information Hidden in Gradients of Regression with Target Noise

Arash Jamshidi, Katsiaryna Haitsiukevich, and Kai Puolamäki.

AISTATS 2026. PDF

[2] GRADSTOP: Early Stopping of Gradient Descent via Posterior Sampling

Arash Jamshidi, Lauri Seppäläinen, Katsiaryna Haitsiukevich, Hoang Phuc Hau Luu, Anton Björklund, and Kai Puolamäki.

ECAI 2025. PDF

[3] Extended Deep Submodular Functions

Seyed Mohammad Hosseini*, Arash Jamshidi*, Seyed Mahdi Nourmousavi, Mahdi Jafari Siavoshani, and Naeimeh Omidvar.

Transactions on Machine Learning Research, 2024. PDF

[4] Differentially Private Machine Learning-Powered Combinatorial Auction Design

Arash Jamshidi*, Seyed Mohammad Hosseini*, Seyed Mahdi Nourmousavi, and Mahdi Jafari Siavoshani.

arXiv preprint, 2024. PDF

* Equal contribution

Education

ELLIS Ph.D. Program, Ph.D. in Computer Science, 2028 (expected).
Primary host: University of Helsinki. Exchange host: KTH Royal Institute of Technology.

Sharif University of Technology, M.Sc. in Artificial Intelligence and Robotics, 2024.

Honors

Received Ph.D. offers from multiple ELLIS PIs for the ELLIS Ph.D. Program (acceptance rate ≈ 3%), 2024.
Ranked 10th in the Nationwide M.Sc Entrance Exam in Computer Engineering in Iran among more than 16,000 participants (top 0.06%), 2021.

Courses

Theoretical and advanced courses from my graduate studies include:

Measure Theory
Fourier Analysis
High-Dimensional Probability
Machine Learning Theory
Stochastic Processes
Information Theory and Coding
Convex Optimization
Convex Analysis II
Real Analysis
Information-Theoretic Methods in High-Dimensional Statistics
Advanced Linear Algebra
Approximation Algorithms
Online Learning and Bandits
Algorithmic Game Theory
Submodular Functions and Optimization
Algorithms for Big Data