Research Interests
- Machine Learning (ML) & Deep Learning (DL)
- Explainable AI (XAI)
- Reinforcement Learning (RL)
- Physics-Informed Neural Network (PINN)
- Natural Language Processing (NLP)
- Quantum Computing
- Quantum Machine Learning (QML)
- Uncertainty Quantification
- Comparative Genomics
Research Experiences
| Visiting Researcher (VR) | Mar 2024 - Present |
|---|---|
| Okinawa Institute of Science and Technology Graduate University (OIST), Japan | Affiliation |
| Research topic: Evolutionary dynamics of strongly conserved sequences in vertebrates and insects | Code Repository |
| Supervisor: Professor Jonathan Miller (BS in Statistics, Yale; PhD in Biology, Cambridge; PhD in Physics, Caltech) | |
| ■ Contributed to FY2023 Annual Report |
| Lead Researcher | Mar 2023 - Present |
|---|---|
| Advanced Machine Intelligence Research Lab (AMIRL), American International University-Bangladesh (AIUB), Bangladesh | Portfolio |
| Roles: Research Assistant (Mar 2023 - Dec 2023), Researcher (Dec 2023 - Feb 2024), Lead Researcher (May 2024 - Present) | |
| Research Affiliations: | |
| Department of Natural Language Processing and Computational Linguistics | |
| Supervisor: Dr. Mohammad Firoz Mridha (Associate Professor, Department of Computer Science, AIUB) | |
| Collaborators: Prof. R. Simon Sherratt (IEEE Fellow), Prof. Nilanjan Dey, Prof. Jungpil Shin, Prof. Yuichi Okuyama, Prof. Zeyar Aung, Prof. Yutaka Watanobe, Prof. Md. Rashedul Islam | |
| ■ Published articles in the fields of DL, QML, GNN, XAI, conformal prediction, human-in-the-loop, NLP, and operations research |
| Visiting Research Student (VRS) | Feb 2023 - Feb 2024 |
|---|---|
| Okinawa Institute of Science and Technology Graduate University (OIST), Japan | Affiliation |
| Research topic: Evolutionary dynamics of strongly conserved sequences in vertebrates and insects | Code Repository |
| Supervisor: Professor Jonathan Miller (BS in Statistics, Yale; PhD in Biology, Cambridge; PhD in Physics, Caltech) | |
| Collaborators: Dr. Lucia Zifcakova, Dr. Priscila Do Nascimento Biller, Dr. Zdenek Lajbner, and Dr. Reuven Pnini | |
| ■ Critically analyzed and visually represented all potential combinations of inter-gap segments (IGS), ancestral repeats (ARs), and contiguous mismatched ARs in human/mouse and human/gorilla genome alignments, focusing on both DNAs and repetitive sequences | |
| ■ Successfully replicated the findings of the neutral indel model proposed by Lunter, Pointing, and Hein (2006) |
| Research Lead | May 2022 - Mar 2023 |
|---|---|
| Research Camp 02, Scholarship School BD, Bangladesh | |
| Research topic: Ultrasound-Based AI for COVID-19 Detection | |
| Supervisor: Dr. Mohammad Arafat Hussain (Post-doctoral Research Fellow at Image, Informatics & Intelligence Research Lab, Harvard Medical School; PhD in Biomedical Eng., UBC Canada; MASc in Biomedical Eng., UBC) | |
| Experiences: | |
| ■ Led the research team of 17 fellow researchers on a project titled “Ultrasound-Based AI for COVID-19 Detection: A Comprehensive Review of Public and Private Lung Ultrasound Datasets and Studies” | |
| ■ Contributed to writing the original manuscript, software implementation, and data curation and served as a corresponding author for the entire communication with the journal |
| Research Intern (RI) | Oct 2021 - Mar 2022 |
|---|---|
| Okinawa Institute of Science and Technology Graduate University (OIST), Japan | Affiliation (Acceptance Rate: 14%) |
| Research topic: Evolutionary dynamics of strongly conserved sequences in vertebrates and insects | Code Repository |
| Supervisor: Professor Jonathan Miller (BS in Statistics, Yale; PhD in Biology, Cambridge; PhD in Physics, Caltech) | |
| Experiences: | |
| ■ Awarded with full-funded scholarship with a daily allowance of JPY 2400 per working day (taxable) | |
| ■ Conducted research on perfectly conserved sequence (PCS) length distributions of UCSC 44 pairwise genome sequences | |
| ■ Analyzed quantile kurtosis of PCS lengths proposed by Nash and Lenhard (2018) and identified a ‘knee’ in the PCS distributions of the heavy-tailed region | |
| ■ Collaborated with Dr. Zifcakova to track erroneous out-of-bound PCS coordinates generated by Nash and Lenhard (2018), utilizing R and Bedtool, and resolved issues using Python 3.10 | |
| ■ Optimized Nash and Lenhard’s 3 R scripts for PCS generation, quantile kurtosis analysis, and genomic regulatory blocks (GRBs), reducing time and memory complexity. Successfully reproduced PCS coordinates following UCSC format and fixed genome coordinate related errors in R’s Bioconductor package | |
| ■ Demonstrated proficiency in shell scripting, parallel computing, and software installation on the HPC Deigo cluster | |
| ■ Presented weekly progress to the Unit using MS PowerPoint slides and animations |
| Research Intern | May 2021 |
|---|---|
| UiT - The Arctic University of Norway | Internship Report |
| Supervisors: Ghada Bouzidi, Prof. Aleksander Pedersen, Prof. Rune Dalmo | |
| Duties included: | Code Repository |
| ■ Conducted comprehensive data and statistical analysis on the Narvik road dataset as part of the DIT4BEARs Smart Road project | |
| ■ Designed, implemented, and evaluated ML models (classifiers: KNN, Logistic Regression, XGBoost, Prophet) that identified the road state as one of the following states: dry, moist, wet, icy, snowy, slushy) to tackle the winter weather challenges encountered in the Barren Euro-Arctic region | |
| ■ Proposed a novel safety metric and forecasted its values using ML regressors: Ridge, Lasso, Elastic Net, Linear Regression, XGBRegressor | |
| ■ Effectively communicated project findings by presenting results at the final meeting, attended by 4 distinguished professors |
Grants & Funding
Competitive Research Funding in AI
-
University of Aizu, Japan 2023
QAmplifyNet: pushing the boundaries of supply chain backorder prediction using interpretable hybrid quantum-classical neural network
Research Sponsor: Prof. Jungpil Shin -
University of Aizu, Japan 2023–Present
Analyzing Male Domestic Violence through Exploratory Data Analysis and Explainable Machine Learning Insights
Research Sponsor: Prof. Jungpil Shin -
Khalifa University, UAE 2024–Present
TriQXNet: Forecasting Dst Index from Solar Wind Data Using an Interpretable Parallel Classical–Quantum Framework with Uncertainty Quantification
Research Sponsor: Prof. Zeyar Aung -
Khalifa University, UAE 2024–Present
KACQ-DCNN: Uncertainty-Aware Interpretable Kolmogorov-Arnold Classical-Quantum Dual-Channel Neural Network for Heart Disease Detection
Research Sponsor: Prof. Zeyar Aung -
Okinawa Institute of Science and Technology (OIST), Japan 2024–Present
MCDFN: Supply Chain Demand Forecasting via an Explainable Multi-Channel Data Fusion Network Model
Role: Visiting Researcher -
University of Aizu, Japan 2024
A Hybrid Transformer and Attention-Based Recurrent Neural Network for Robust and Interpretable Sentiment Analysis of Tweets
Research Sponsor: Prof. Yuichi Okuyama
Peer Review Experiences
Reviewed a total of 16 journal articles verified by Web of Science, as of December 2024 (Web of Science ResearcherID: AFE-0313-2022).
- IEEE Access
- Computers & Industrial Engineering (Elsevier)
- Expert Systems with Applications (Elsevier)
- Multimedia Tools and Applications (Springer Nature)
- Engineering Applications of Artificial Intelligence (Elsevier)
- The Journal of Supercomputing (Springer Nature)
- Cluster Computing (Springer Nature)
- Scientific Reports (Nature Portfolio)
- Journal of Contemporary African Studies (Taylor & Francis)
- Journal of Multidisciplinary Healthcare (Taylor & Francis)
- 7th European Conference on Industrial Engineering and Operations Management (Augsburg, Germany, July 2024)
Leadership and Advisory Roles
| KUET Research Society | Nov 2023 - May 2024 |
|---|---|
| Co-founder & President | LinkedIn Page |
| Co-moderator: Dr. Md. Saiful Islam | |
| Responsibilities: | |
| ■ Served as an Executive Committee Member and President of the Industrial Engineering and Management Research Unit | |
| ■ Supervising (as an alumnus) 6 groups of research students concentrating on ML-DL and computational fuzzy logic, fostering cross-departmental research collaboration | |
| ■ Teaching (as an alumnus) scientific research methodology, research ethics, and journal article formatting meeting publication criteria and acquiring funding for publications |
| SpecX | Aug 2022 - Present |
|---|---|
| Founder & AI Researcher | MIT Solve 2024 |
| ■ Developed a hybrid DL model architecture, integrated XAI, topic modeling, telemedicine appointment, and personalized treatment | |
| ■ Acquired funding from the University of Aizu (Japan) and AIUB (Bangladesh) to collect patients’ data and conduct NLP research |