Abu Zaher Md. Faridee, Adjunct Assistant Professor
Dr. Faridee’s professional work and research focus on constructing scalable machine learning models resilient against domain and category shifts with minimal-to-no additional supervision. He primarily engages with deep domain adaptation, unsupervised, self-supervised, adversarial, disentangled representation learning, novel category discovery, and learnable data augmentation techniques, with practical applications in text, audio, and video domains. Dr. Faridee aims to discover the optimal transferability of representations across domains, tasks, and modalities, addressing real-world ML challenges with these methodologies.
At Amazon, Dr. Faridee’s team developed an end-to-end neural machine translation pipeline enhancing Amazon’s customer service experience. He enhanced the robustness of NMT models against noisy, out-of-domain inputs using the aforementioned approaches. In 2021, he interned with Microsoft Research’s Audio and Acoustics Research Group, devising novel deep neural architectures for estimating deep noise suppression model performance.
Before obtaining his Ph.D., Dr. Faridee spent approximately 8 years constructing, assembling, and leading teams in distributed data processing and analytics back-ends, catering to millions of users in the USA and UK. From 2009–2013, he actively contributed to several open-source NLP and ML projects through the Google Summer of Code program, both as a participant and later as a mentor.
LinkedIn: www.linkedin.com/in/azmfaridee
Anuradha Ravi, Research Assistant Professor
Dr. Ravi is a Research Assistant Professor in Information Systems at UMBC. Her research interests include edge computing, developing low-power machine learning models for IoT, IoT-enhanced smart spaces, and wireless networking. Prior to joining UMBC, Dr. Ravi worked as a Research Scientist (PostDoc) at Singapore Management University (Oct 2018 – Mar 2023). She completed her Ph.D. at the Indian Institute of Technology, Roorkee, India, in 2016. Dr. Ravi was awarded the “SMU Research Excellence” award for her contribution to the “AI-Based Occupancy Aware Smart Building Energy Optimization project,” which includes indoor localization for occupancy sensing in Smart Buildings. She has vast experience building real-time working prototypes to convert research into real-time demonstration.
Personal Website: https://sites.google.com/view/anuradhar
LinkedIn: https://www.linkedin.com/in/anuradha-ravi-ph-d-36564a3b/
Catherine Ordun, Ph.D. ’23, information systems, Adjunct Assistant Professor
Catherine Ordun is a Vice President at Booz Allen Hamilton leading AI Rapid Prototyping in the Office of the CTO. She graduated in 2023 with a PhD from the Department of Information Systems under Dr. Sanjay Purushotham and Co-advised by Dr. Edward Raff. Her dissertation was entitled “Multimodal Deep Generative Models for Cross Spectral Image Analysis.” Her research interests combine industry innovation and AI development in Multimodal AI, biometrics, and visible-thermal image registration. At Booz Allen, she coordinates with leaders in defense, civil, and health as an AI SME to develop novel and scalable technical approaches. She has also supported the Mark Cuban AI Bootcamp since 2020, which trains high school students to learn the basics of AI. In her spare time, she writes science fiction, runs, and spends time with her family.
LinkedIn: https://www.linkedin.com/in/catherine-ordun
Nurcan Tufekci, Faculty Research Assistant
Ph.D. Candidate Tufekci is a Faculty Research Assistant in the Department of Information Systems at UMBC. Her educational background includes a Bachelor’s degree in Family and Consumer Sciences, a Master’s degree in Women’s Studies, and a Master’s in Family and Consumer Sciences. She is currently pursuing a Ph.D. in Sociology. Her research interests include women, family, migration, the virtual world, and its impact on society. She has field experience with migrants. For her master’s thesis, she conducted face-to-face interviews with 200 Syrian women in their homes. She has publications from her thesis, including peer-reviewed publications. She also published her study on Turkish immigrants living in Germany in a peer-reviewed journal, for which she collected field data online from Turkey. She contributed to research on the analysis of female surgeons in Turkey. She participated in the field data for this research. Her interests include children’s fairy tales and documentary photography.
Philip Feldman, M.S. ’14, human-centered computing, and Ph.D. ’20, human-centered computing, Adjunct Assistant Professor
Dr. Feldman has been an Adjunct Research Assistant Professor of Information Systems at UMBC since 2020. He has a diverse and extensive professional background, having served as an AI/ML Futurist at ASRC Federal, Technology Architect at Novetta, Inc, and in various technical and leadership roles at other organizations. His educational background includes a PhD and MS in Human-Centered Computing from UMBC, an MS in Interdisciplinary Studies from Johns Hopkins University, and a BA in Interdisciplinary Studies from the University of Maryland. He has been a speaker at several prominent events and has authored the book “Stampede Theory: Human Nature, Technology, and Runaway Social Realities” published by Elsevier in 2023. Additionally, he has a rich history of publications, speaking engagements, and patents in the fields of AI, machine learning, and human-centered computing. His research interests include societal-scale AI conflict, intelligent control systems, simulation, ethics, and the impact of technology on society.
Personal Website: https://philfeldman.com
LinkedIn: https://www.linkedin.com/in/phil-feldman-phd/