Assistant Professor (UK Lecturer)
Lancaster University
I am an Assistant Professor (UK Lecturer) in the School of Mathematical Sciences at Lancaster University, within the MARS (Mathematics for AI in Real-world Systems) section. I am also a member of Data Science and AI @ Lancaster (DSAIL) and the European Laboratory for Learning and Intelligent Systems (ELLIS).
I was a postdoctoral researcher at the Institute of Science and Technology Austria (ISTA), working in the group led by Prof. Dan Alistarh. I was awarded a Marie Skłodowska-Curie Fellowship through the MSCA COFUND IST-BRIDGE program at ISTA. As part of the fellowship, I had the opportunity to work with Dr. Alexandre Marques during an industrial secondment at Neural Magic, Inc. (now acquired by Red Hat) in the USA. Before joining ISTA in 2022, I was a postdoctoral research fellow at King Abdullah University of Science and Technology (KAUST) in Saudi Arabia from 2019 to 2022, in the group led by Prof. Peter Richtárik. Prior to that, I worked with Prof. Diogo Gomes at KAUST as a research technician from 2016 to 2019. I obtained my Ph.D. in Mathematics in 2018 from Yerevan State University (YSU) in Armenia, under the supervision of Prof. Grigori Karagulyan.
Open Position: I am looking for a highly motivated PhD student to work on Optimization Theory and Algorithms for Machine Learning. The position is open to applicants with UK fee status. Details of this project, along with the full list of available PhD projects within MARS (Mathematics for AI in Real-world Systems), can be found here. Before submitting a formal application, interested candidates should contact me with their CV, academic transcript/grades, and a statement describing motivation and research interests.
My research focuses on optimization theory and algorithms for machine learning, with an emphasis on efficiency, scalability, and the theoretical understanding of optimization methods. These methods are particularly relevant for large-scale machine learning training and federated learning.
I completed my Ph.D. in real harmonic analysis, a branch of mathematics that explores the relationship between functions or signals and their frequency domain representations. My thesis investigated the convergence and divergence properties of certain convolution-type integral operators. In addition, I have done some research in algebra. During my undergraduate studies at YSU, I completed a research project on universal algebraic structures called dimonoids, which led to a publication in Algebra and Discrete Mathematics. Later, at KAUST, I worked on symbolic computation, specifically on developing computer algebra techniques for automating certain aspects of PDE analyses.
For the complete list of my publications, please visit my Google Scholar page.
Honoured to be recognised as a Top Area Chair (Main Track) at NeurIPS 2026. Many thanks to the Program Committee for the recognition and their efforts throughout the review process.
Together with Matthias Sachs and Maciej Buze, I organised the inaugural MARS Annual Symposium 2026, held at Lancaster University on 9–11 September 2026. The symposium brought together researchers working on computational mathematics and probabilistic machine learning.
How Do We Cook Up Artificial Intelligence With Optimization Algorithms? — an accessible introduction to how optimization algorithms train AI models, explained through the analogy of developing and perfecting a cooking recipe. Joint work with Eugenia Iofinova and Dan Alistarh.
MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning - joint work with Ionut-Vlad Modoranu and Dan Alistarh.
LoRDO: Distributed Low-Rank Optimization with Infrequent Communication - joint work with Andrej Jovanović, Alex Iacob, Ionut-Vlad Modoranu, Lorenzo Sani, William F. Shen, Xinchi Qiu, Dan Alistarh, Nicholas D. Lane.
DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers - joint work with Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis, Dan Alistarh.
Towards Robust Scaling Laws for Optimizers - joint work with Alexandra Volkova, Christoph H. Lampert, Dan Alistarh.
LoRDO: Distributed Low-Rank Optimization with Infrequent Communication - joint work with Andrej Jovanović, Alex Iacob, Ionut-Vlad Modoranu, Lorenzo Sani, William F. Shen, Xinchi Qiu, Dan Alistarh, Nicholas D. Lane.
DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers - joint work with Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis, Dan Alistarh.
CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training - joint work with Soroush Tabesh, Andrei Panferov, Alexandra Volkova, Dan Alistarh.
FFT-based Dynamic Subspace Selection for Low-Rank Adaptive Optimization of Large Language Models - joint work with Ionut-Vlad Modoranu, Erik Schultheis, Max Ryabinin, Artem Chumachenko, Dan Alistarh.
MT-DAO: Multi-Timescale Distributed Adaptive Optimizers with Local Updates - joint work with Alex Iacob, Andrej Jovanovic, Meghdad Kurmanji, Lorenzo Sani, Samuel Horváth, William F. Shen, Xinchi Qiu, Nicholas D. Lane.
DES-LOC: Desynced Low Communication Adaptive Optimizers for Training Foundation Models - joint work with Alex Iacob, Lorenzo Sani, Paris Giampouras, Samuel Horváth, Andrej Jovanovic, Meghdad Kurmanji, Preslav Aleksandrov, William F. Shen, Xinchi Qiu, Nicholas D. Lane.
I'm excited to announce that I've started my new position as an Assistant Professor (UK Lecturer) in the School of Mathematical Sciences at Lancaster University, joining the MARS (Mathematics for AI in Real-world Systems) section.
CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training - joint work with Soroush Tabesh, Andrei Panferov, Alexandra Volkova, Dan Alistarh.
MT-DAO: Multi-Timescale Distributed Adaptive Optimizers with Local Updates - joint work with Alex Iacob, Andrej Jovanovic, Meghdad Kurmanji, Lorenzo Sani, Samuel Horváth, William F. Shen, Xinchi Qiu, Nicholas D. Lane.
Unified Scaling Laws for Compressed Representations - joint work with Andrei Panferov, Alexandra Volkova, Ionut-Vlad Modoranu, Vage Egiazarian, Dan Alistarh.
GradSkip: Communication-Accelerated Local Gradient Methods with Better Computational Complexity - joint work with Artavazd Maranjyan and Peter Richtárik.
Unified Scaling Laws for Compressed Representations - joint work with Andrei Panferov, Alexandra Volkova, Ionut-Vlad Modoranu, Vage Egiazarian, Dan Alistarh.
DES-LOC: Desynced Low Communication Adaptive Optimizers for Training Foundation Models - joint work with Alex Iacob, Lorenzo Sani, Paris Giampouras, Samuel Horváth, Andrej Jovanovic, Meghdad Kurmanji, Preslav Aleksandrov, William F. Shen, Xinchi Qiu, Nicholas D. Lane.
SVD-Free Low-Rank Adaptive Gradient Optimization for Large Language Models - joint work with Ionut-Vlad Modoranu, Erik Schultheis, Dan Alistarh.
LDAdam: Adaptive Optimization from Low-dimensional Gradient Statistics - joint work with Thomas Robert, Ionut-Vlad Modoranu, Dan Alistarh.
LDAdam: Adaptive Optimization from Low-dimensional Gradient Statistics - joint work with Thomas Robert, Ionut-Vlad Modoranu, Dan Alistarh.
As part of the fellowship, I have started 6-month secondment with Neural Magic, Inc. in the USA, working with Dr. Alexandre Marques on LLM quantization.
MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence - joint work with Ionut-Vlad Modoranu, Grigory Malinovsky, Eldar Kurtic, Thomas Robert, Peter Richtárik, Dan Alistarh.
The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information - joint work with Diyuan Wu, Ionut-Vlad Modoranu, Denis Kuznedelev, Dan Alistarh.
