Asif Mohammed Saad

Beyond Classical Limits: How Quantum SVMs and Feature Maps Revolutionize Machine Learning

This paper was written with A. M. Saad, K. E. Rahman, M. F. Shahriyar, Y. Gulora, and H. M. Hossain.

It appeared in the Proceedings of the International Conference on Big Data, IoT and Machine Learning, pages 619 to 630, in 2025.

SegFormer Fine-Tuning with Dropout: Advancing Hair Artifact Removal in Skin Lesion Analysis

This paper was written with A. M. Saad and U. N. Mahi.

It was posted to arXiv as preprint 2509.02156 in September 2025.

arXiv PDF

Evaluating YOLO Architectures: Implications for Real-Time Vehicle Detection in Urban Environments of Bangladesh

This paper was written with H. M. Hossain, P. Nath, M. N. Mahi, I. Uddin, I. J. Eiste, S. N. R. Ratul, M. N. U. Mozumdar, A. M. Saad, and M. T. Hossain.

It was posted to arXiv as preprint 2509.05652 in September 2025.

arXiv PDF

Assessment of YOLO Models for Real-Time Vehicle Detection in the Complex Urban Settings of Bangladesh

This paper was written with H. M. Hossain, P. Nath, M. I. Uddin, I. J. Eiste, M. N. Mahi, S. N. R. Ratul, M. N. U. Mozumdar, A. M. Saad, and M. T. Hossain.

It appeared in the Proceedings of the 28th International Conference on Computer and Information Technology in 2025.

Bangla News Article Dataset

This paper was written with A. M. Saad, U. N. Mahi, M. S. Salim, and S. I. Hossain.

It was published in Data in Brief, volume 57, article 110874, in 2024.

The paper introduces a public corpus of Bangla news articles that supports text classification, summarization, and language modeling research in a low-resource language.

DOI

Forecasting Crucial Biogeochemical Indicators of the Southern Ocean for Climate Monitoring Using Modified Kernel-Based Support Vector Regression

This paper was written with A. M. Saad, R. Mahmud, and S. Das.

It appeared in the Proceedings of the International Conference on Big Data, IoT and Machine Learning, pages 809 to 823, in 2023.

Autonomous Navigation in Bangladesh: A Multi-Camera Perception-to-Action Approach

This project is funded by a grant from the Institute of Research and Publications at Premier University, Chattogram, awarded in 2026.

I am building a multi-camera perception-to-action pipeline for road conditions in which dense traffic and unstructured lane behavior break the assumptions built into systems developed elsewhere.

MNIST Classification with Quantum Neural Networks

I built this project using PyTorch and the Classiq SDK.

It is a hybrid quantum neural network for digit classification, and it covers the full path from data encoding through circuit design, training, and evaluation.

Code

Quantum Algorithms Notebooks

These notebooks cover variational algorithms and quantum simulation.

I implemented quantum data encoding, qubit mapping, variational quantum eigensolvers, and quantum phase estimation, and documented each one as I worked through it.

Code