Amjad Seyedi
 |
Doctoral researcher |
๐ค About me
I am a Doctoral Researcher at the University of Mons, supervised by Prof. Nicolas Gillis. My current research sits at the intersection of Matrix Theory and Optimization, specifically developing matrix factorization and low-rank approximation methods for Trustworthy Machine Learning. I aim to leverage these mathematical tools to enhance robustness, interpretability, and fairness in high-dimensional data and network analysis. Previously, I was a Graduate Research Assistant at the University of Kurdistan, where I led the Algebraic Machine Learning (AML) team and focused on representation learning. I hold an MSc in Artificial Intelligence from the same institution, where my work with Dr. Parham Moradi and Dr. Fardin Akhlaghian focused on matrix factorization for semi-supervised learning and recommender systems. I also hold a Bachelorโs degree in Software Engineering.
๐ Research interests
๐ง Machine Learning: representation learning, deep learning, unsupervised learning
๐ก๏ธ Trustworthy ML: robustness, generalization, interpretability, fairness
๐งฎ Applied Mathematics: linear algebra, optimization, low-rank approximation
๐ Applications: healthcare, recommender systems, remote sensing
๐ข News
Upcoming
๐๏ธ I will attend the 2026 European Signal Processing Conference (EUSIPCO), Bruges, Belgium, 31 August โ 4 September, 2026.
๐๏ธ I will attend the 2026 Foundations of Computational Mathematics (FoCM), Vienna, Austria, July 16-18, 2026.
๐ I will present my PhD confirmation in Matrix Theory and Optimization at Polytech MONS, University of Mons, Belgium, June 24, 2026.
Recent news
๐ฐ Our paper, Bilinear Nonnegative Matrix Factorization for Hyperspectral Unmixing, has been accepted for the European Signal Processing Conference 2026.
๐๏ธ I attended the 2026 International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Barcelona, Spain, May 4-8, 2026.
๐ฐ Our paper, Robust Asymmetric Encoder-Decoder Nonnegative Matrix Factorization for Hyperspectral Anomaly Detection, has been accepted for publication in Neurocomputing 2026.
๐ฐ Our paper, Robust log-based multi-label feature selection with dynamic label correlation and relevanceโredundancy optimization, has been accepted for publication in Knowledge-Based Systems 2026.
๐ฐ Our paper, Contrastive Calibration on Consensus and Complementary Multi-View Representations, has been accepted for publication in Pattern Recognition 2026.
๐ฐ Our paper, Encoder-Decoder Symmetric Nonnegative Matrix Tri-Factorization for Graph Clustering, has been accepted for the International Conference on Acoustics, Speech, and Signal Processing 2026.
๐ฐ Our paper, Semantic Encoder-Decoder NMF with Kullback-Leibler Divergence, has been accepted for publication in International Journal of Machine Learning and Cybernetics 2026.
๐ฐ Our paper, Community Detection via Deep Motif-regularized Asymmetric NMF, has been accepted for publication in , Engineering Applications of Artificial Intelligence 2026.
2025
๐๏ธ I attended the “Autumn School: Scientific Machine Learning and Numerical Methods”, at the Centrum Wiskunde & Informatica (CWI), Netherlands, October 27-31, 2025.
๐ฐ Our paper, “A Deep Latent Factor Graph Clustering with Fairness-Utility Trade-off Perspective”, has been accepted for the 2025 IEEE International Conference on Big Data.
๐ฐ Our paper, Distributionally Robust NMF with Self-Paced Learning, has been accepted for publication in Information Sciences journal 2026.
๐ My three Master's students successfully defended their theses in Artificial Intelligence at the Algebraic ML Team, University of Kurdistan, September-October 2025.
๐๏ธ I attended the “SOCN: Stochastic variance-reduced optimization algorithms and applications to federated learning” course, UCLouvain, Belgium, October 7-9, 2025.
๐๏ธ I attended the Systems, Optimization, Control and Networks (SOCN) Workshop, UMONS, Belgium, September 26, 2025.
๐๏ธ I was on the organizing committee of the third workshop on Low-Rank Models and Applications (LRMA 25), Mons, September 11-12, 2025.
๐๏ธ I attended the “PhD School: Machine Learning and Optimization”, at the Centrum Wiskunde & Informatica (CWI), Netherlands, September 2-4, 2025.
๐ฐ Our paper, Encoder-Decoder NMF with ฮฒ-Divergence, has been accepted for publication in Pattern Recognition journal 2026.
๐ฐ Our paper, Instance-wise Distributionally Robust NMF, has been accepted for publication in Pattern Recognition journal 2026. It's my first collaboration with Prof. Nicolas Gillis!
๐๏ธ I attended the “Learning Theory from First Principles” course (Prof. Francis Bach), KU Leuven, Belgium, April 1-3, 2025.
๐๏ธ I attended the Masterclass: “Graphs, Data and AI” (Prof. Jure Leskovec), University of Antwerp, Belgium, March 20, 2025.
๐ค๏ธ I delivered a speech on Deep Matrix Factorizations for Data Representation at the STADIUS Center, KU Leuven, Belgium, February 20, 2025.
2024
๐๏ธ I attended the Graduate School in Systems, Optimization, Control and Networks (SOCN) Study Day, UCLouvain, Belgium, October 29, 2024.
๐๏ธ I attended the Autumn School on Constrained Optimization and Machine Learning, Trier University, Germany, October 9-11, 2024.
๐๏ธ I attended the ALGOPT2024 Workshop on Algorithmic Optimization: Tools for AI and Data Science, UCLouvain, Belgium, August 27-30, 2024.
๐๏ธ My poster was accepted for presentation at the ALGOPT2024 Workshop on Algorithmic Optimization: Tools for AI and Data Science.
๐ฑ I started my PhD in Matrix Theory and Optimization at Polytech MONS, University of Mons, Belgium, June 18, 2024.