Laboratory of Intelligent Learning Agents and Algorithms

Computer Science Department
University of Saskatchewan

A fundamental question to understanding intelligence is how knowledge can be generalized and reused when making decisions to solve complex tasks. The overarching goal of our research is to design new reinforcement learning algorithms that effectively learn a reusable knowledge base from experience. Our approaches include designing algorithms that learn representations or abstractions from experiences to improve their performance. Successfully building such algorithms expands our algorithmic understanding of intelligence.