We study artificial intelligence & workflow automation
Research Statement
Fundamental AI research
We aim to advance the foundations of data science, machine learning, and agentic AI.
- mathematical models of neural network dynamics
- meta-learning methods for ensemble learning
- integration of AI with classical computer science
Novel applications of AI
Our research focuses on discovering and implementing novel applications of AI technology, particularly in workflow automation. We aim to enhance efficiency, reduce costs, and improve accuracy in various business processes.
Open-source benchmarks
We are committed to providing the research community with open source benchmarks to evaluate AI models and technologies. These benchmarks are crucial for ensuring transparency, reproducibility, and progress in the field of AI research.
AI readiness report for industry sectors
We assess AI readiness across various industry sectors, identifying opportunities and challenges. Our reports offer insights into the practical implementation of AI technologies and their impact on industry operations.
Members
Ken Pu is an associate professor in Computer Science at Ontario Tech University. He has been actively working in the intersection of database systems, text and natural language processing, machine learning and artificial intelligence. He received his PhD in Computer Science from University of Toronto in 2006.
Limin Ma is working on AI agents and agentic workflows involving multimodal and custom LLMs. He obtained his Masters in Computer Science from Ontario Tech University where he worked on embedded neural networks for keyword query optimization.
Zikun Fu is working on instruction tuned embedding models for semantic retrieval applications. He is currently working on benchmarks for instruction based embedding models. Zikun is currently working on his master’s degree.
Chen Yang is a visiting researcher for the duration of Summer 2024. Our collaboration is on studying the embedding models with instructions: their evaluation and optimization. Chen is completing his master’s degree from Northeasstern University.
Projects
Relational Data Management Using Vector Space Approach
We are investigating various learning based approaches to a range of database problems in the intersection of database Management and natural language processing. By leveraging large language models (Minaee et al. 2024) and tabular embedding methods (Singh and Bedathur 2023), we are working on new information retrieval techniques.
AI Applications Development Frameworks
AI has created new opportunities in application design. However, effective AI application development comes with many interesting challenges. Despite numerous libraires (LlamaIndex 2024),(LangChain 2024), (Khattab et al. 2023) there are many unsolved issues and confusions faced by developers. We are exploring the design space for the next generation of AI application development framework by combining time proven practices in Web application, and server development experiences of the last several decades.