image

I’m a PhD student in Computer Science at the University of Kentucky, where I work at the intersection of deep learning, multimodal learning, and real-world data challenges. My research focuses on building effective and efficient AI solutions that can adapt to diverse domains. I earned my BSc in Computer Science and Engineering from RUET and previously served as a lecturer at Green University of Bangladesh. As a Kaggle Competition Expert, I enjoy solving practical machine learning problems and learning from the global data science community. I’m interested in new ideas and exciting collaborations!

RECENT NEWS

CausalGeD: Blending Causality and Diffusion for Spatial Gene Expression Generation
May 2025

Accepted at KDD 2025! (Acceptance rate < 19%)

Novel blending of diffusion and AR in Spatial Transcriptomics. Long-range dependencies in gene expression. Achieves SOTA results on 10 datasets.
TSCMamba: Mamba meets multi-view learning for time series classification
March 2025

Accepted at Information Fusion! (Impact Factor 14+)

Robust time-series classification using multi-view learning. Temporal and spectral features for shift-equivariance. Superior accuracy and efficiency over 30 benchmark datasets.
Mol-CADiff: Causality-Aware Autoregressive Diffusion for Molecule Generation
March 2025

Preprint available at arXiv

GraphMinNet: Learning Dependencies in Graphs with Light Complexity Minimal Architecture
February 2025

Preprint available at arXiv

Low-dose computed tomography perceptual image quality assessment
January 2025

Accepted at Medical Image Analysis! (Impact Factor 10+)

First low-dose CT IQA challenge. First open-source low-dose CT IQA dataset. Potential of no-reference IQA methods.
Timemachine: A time series is worth 4 mambas for long-term forecasting
July 2024

Accepted at 27th ECAI (Acceptance rate < 24%)

Unifies channel-mixing and channel-independence. Superior performance in prediction accuracy. Small memory frootprint.
MambaTab: A Plug-and-Play Model for Learning Tabular Data
May 2024

Accepted at 7th IEEE MIPR (Acceptance rate < 20%)

Extremely small model size. Linear scalability. Effective end-to-end training and inference.

Contact

Get in Touch

Get in Touch

Kindly drop me a text with the following email (remove *)

Lexington, KY, USA
atik*ahamed*[at]uky.edu