My name is Md Mahabub Uz Zaman. I’m a doctoral student in Computer Science at Texas Tech University. I’m interested in multi-agent systems, especially how rational agents interact with each other. I want to design or generalize rational agents that can adapt in cooperative or adversarial settings using Reinforcement Learning and knowledge from Psychology, Social Science, and Economics.
Another area of my research interest is machine unlearning. My goal is to enhance the robustness and privacy of models, particularly in healthcare, where privacy is paramount, and to mitigate cyber threats. I am actively engaged in research and exploration within this field. Furthermore, I am particularly enthusiastic about multi-modal learning in healthcare. I am currently working on multi-modal learning for disease prediction and progression. By integrating diverse data types, we can enhance healthcare decision-making and improve the quality of care. Prior to my doctoral program, I obtained a Master’s degree in Industrial Engineering and Management, with a specialization in operations research. Additionally, I have engaged in projects related to computer vision and transformer architecture for generative AI. I designed a question-answer model as an intelligent agent. I am also interested in research involving vision and text-based domains.
I'm also fascinated by the intersection of AI and biology, particularly in the fields of aging and multi-omics. I am actively engaged in research in this area, including projects focused on aging mechanisms and multi-omics datasets. I have developed high-accuracy machine learning models to estimate chronological age from mouse brain metabolomic profiles. I have also successfully designed and trained an explainable deep learning model to classify exosomes from multiple mammal species using Raman spectroscopy data. I am also currently working on a transcriptomics project on mouse testes, where I am involved in everything from conducting high-throughput omics experiments to performing detailed RNA-seq data analysis. The opportunity to leverage my skills in deep learning, data analysis, and high-performance computing to uncover the biological secrets of aging and longevity through multi-omics data is something I find incredibly exciting and rewarding.
Email: m[dot]zaman[at]ttu[dot]edu