Admin
Emam Hossain

Ph.D. Student

iHARP: NSF HDR Institute

Causal Artificial Intelligence Lab (CAIL)

University of Maryland Baltimore County

About Me


I am currently pursuing my Ph.D. in the Information Systems Department at the University of Maryland, Baltimore County (UMBC). I’m part of Prof. Md Osman Gani’s Causal Artificial Intelligence Lab (CAIL) at UMBC. My research focuses on machine learning, deep learning, and causality, with a particular emphasis on their applications in climate science. Specifically, I am researching the evolution of supraglacial lakes over the Greenland Ice Sheet using Causal Representation Learning (CRL). My work integrates diverse datasets, such as ICESat-2, CARRA, and ERA5, to uncover the causal impacts of climate variables on lake dynamics like draining, refreezing, and burial.

I have a strong interest in causal methods, including Latent Causal Variable Discovery and multi-source data integration, aiming to enhance model interpretability and align causality across datasets. Additionally, I am exploring how these methods can provide both qualitative and quantitative insights in climate science applications. Beyond this, I have expertise in financial market analysis using time series analysis, such as stock price and foreign exchange rate prediction.

Before starting my Ph.D., I earned my Bachelor's and Master's degrees in Computer Science and Engineering from the University of Chittagong, Bangladesh. My Master’s thesis explored the application of machine learning with Belief Rule-Based Expert Systems for stock price prediction, and a modified version was published in the *Expert Systems with Applications* journal (read here). I have also published several research papers in international journals and conferences (see here).

I worked as a lecturer in the Department of Computer Science and Engineering at Port City International University, Bangladesh, for over three and a half years. I taught core computer science courses, including Artificial Intelligence, Pattern Recognition, Computer Graphics, Theory of Computation, and Object-Oriented Programming. During that time, I supervised multiple undergraduate thesis projects, fostering research skills in students.




Research Interests

My primary research interests lie in machine learning, deep learning, and causal inference, with a focus on their applications in climate science and beyond. Currently, I am working on understanding the causal relationships in climate variables and exploring the use of CRL for analyzing supraglacial lake dynamics. My broader goal is to apply causal methods to uncover latent causal factors shared across multi-source datasets, enabling deeper insights into environmental and scientific phenomena.


Projects


Current Projects

NSF iHARP (May 2022 - Present):
By carefully combining data science and polar research to encourage physics-informed, data-driven discoveries, the iHARP project focuses on increasing our knowledge of the response of polar areas to climate change and its worldwide consequences. iHARP project brought together domain experts from earth science, geology, environmental science, oceanic science, computer science, machine learning, and data science. As a part of the computer science and machine learning community, I’m mainly working on building predictive models based on machine learning and deep learning techniques. Later to improve these predictive models by incorporating causal inference with machine learning to understand the causal structure of the underlying model. This causal structure will help us to determine which parameters have a direct causal effect on sea ice melting and therefore, sea level rise.


Previous Project

NSF Caching-as-a-Service (August 2021 - May 2022):
High-performance, scalable computing system designers have consistently used caching, but since it has been implemented in so many different ways, it may be challenging to standardize and scale in cloud systems. This project develops generalized Caching-as-a-Service (CaaS), elevating the usage of caching in cloud-scale storage systems. A wide range of applications that function in both private and public clouds are supported by the CaaS project. The CaaS project uses use cases from the Cloud, Big Data, and Deep Learning computing paradigms to demonstrate these advances.

Specifically, I worked on identifying different cache workload types (LRU-friendly, LFU-friendly, Scan, and Churn) directly from the trace request sequence. I used FIU’s home trace data from SNIA and developed several machine learning techniques that can successfully identify the abovementioned workload types.