KMA Solaiman (Salvi)

Position

Lecturer

Office

201-C Information Technology and Engineering (ITE) Building

Contact Information

University of Maryland, Baltimore County
1000 Hilltop Circle
Baltimore, MD 21250

email:  ksolaima@umbc.edu
website: https://ksolaiman.github.io/

Education

Ph.D., Computer Science at Purdue University, Summer 2023
M.Sc., Computer Science at Purdue University, 2022
B.Sc., Computer Science and Engineering at Bangladesh University of Engineering and Technology (BUET), 2014

Research Areas

My research interests revolve around Machine Learning and Multimodal Data Management with applications into data discovery, multimodal information retrieval, uncertainty management, recommender systems, natural language processing, and video feature extraction.

Publications

Google Scholar

Biography

I am a full time faculty member and a Lecturer in the Department of Computer Science at University of Maryland Baltimore County (UMBC). My research expertise revolve around multimodal data management and uncertainty management in real-world AI systems using Machine Learning. I completed my Ph.D. from Purdue University, West Lafayette, IN in 2023. I was fortunate to be advised by Professor Bharat Bhargava.

During my time at Purdue, I worked on the REALM project funded by NGC and the SAIL-ON project funded by DARPA. I worked closely with Prof. Michael Stonebraker for the REALM/SKOD project and contributed in scalable Multimodal Information Retrieval and Video and Text Feature Extraction. During the SAIL-ON project, I worked closely with Josh Alspector and rest of the Novelty Working Group (NWG) team on theories of novelty, including novelty detection, adaptation, and characterization in perception and planning domains.

Before coming to Purdue, I served as a full time faculty member at the Ahsanullah University of Science and Technology (AUST) and United International University (UIU) in Dhaka, Bangladesh from 2014 to 2016. My B.Sc. dissertation titled ‘Minimal Parameter Clustering of Complex Shaped Variable-sized Dataset (MPCACS)’ focused on unsupervised learning of complex datasets.

If you are interested in working with me and/or you are already a graduate student at UMBC, feel free to reach out. Undergraduate students can also reach out to me for senior thesis advising or independent work. I also recommend you take at least one of the Machine Learning or NLP or Deep Learning classes before getting into multimodal information retrieval or representation learning.

Recent News Items

Successfully defended his Ph.D. from Purdue University