Dr Sham Puthiya Parambath
- Research Associate (School of Computing Science)
Biography
I, Shameem Ahamed Puthiya Parambath, am a member of the Knowledge & Data Engineering Systems (KDES) under the research group Information, Data and Analysis (IDA) in the School of Computing Science, University of Glasgow, working with Dr Christos Anagnostopoulos and Prof. Roderick Murray-Smith.
I am from a small, beautiful coastal town named Thalassery in the state of Kerala, India, where I spent most of my life. My home is next to the wonderful Government Brennen College, Dharmadom, near a river. I spent two unforgettable years of my life in Brennen.
I completed PhD in Machine Learning from the University of Technology Compiegne (Sorbonne University Association). During that period, I was fortunate to be supervised by Dr. Nicolas Usunier and Dr. Yves Grandvalet. During my Masters at Umea University, I had the fortune of working with Dr. Sihem Amer-Yahia.
Before joining the University of Glasgow, I worked as a Postdoctoral Researcher at Qatar Computing Research Institute, Doha, Qatar. My current work looks at improving sequential algorithms for tasks like dynamic pricing, federated learning, edge computing, etc. At QCRI, I worked in Recommender Systems, Knowledge Graphs and Anomaly Detection. My work on group recommendation and the cold-start recommendation was published in AAAI and ECML conferences. My work on quantifying bias in the knowledge graphs was published in UAI. During my PhD thesis, I worked on multi-objective learning algorithms for multi-class/multi-label classification and personalized recommendations, which was published at venues such as NeurIPS and RecSys.
Research interests
- Learning with feedback
- Mutli-Armed Bandits
- Recommender Systems
- Federated Learning
Google Scholar: [link]
Supervision
- Alfahad, Saleh Abdullah M
Engineering Edge Computing with Predictive Intelligence
Teaching
Database Theory & Applications
Additional information
Reviewer for AAAI 2024, 2021, 2020, 2019; AISTATS 2023, 2022; CIKM 2020; SDM 2020