Levi H. S. Lelis
Appointment
Canada CIFAR AI Chair
Pan-Canadian AI Strategy
About
Appointed Canada CIFAR AI Chair – 2020
Levi Lelis is a Canada CIFAR AI Chair and a fellow at Amii. He is appointed as an assistant professor in the Department of Computing Science at the University of Alberta.
Lelis’ research is dedicated to the development of principled algorithms to solve combinatorial search problems. These problems are integral to optimizing tasks in various sectors. His research group is focused on combinatorial search problems arising from the search for programmatic solutions—computer programs written in a domain-specific language encoding problem solutions. He believes that the most promising path to creating agents that learn continually, efficiently, and safely is to represent the agents’ knowledge programmatically. While programmatic representations offer many advantages, including modularity and reusability, they present a significant challenge: the need to search over large, non-differentiable spaces not suited for gradient descent methods. Addressing this challenge is the current focus of his work.
Awards
- Distinguished Paper Award, IJCAI, 2023.
- Outstanding Reviewer Award, NeurIPS, 2021.
- Distinguished Program Committee Member, IJCAI, 2018-2019
- Outstanding Contribution in Reviewing, Artificial Intelligence journal, 2017
- Winner of the MicroRTS Competition, IEEE, 2018
- Runner-up for the International Planning Competition (IPC), ICAPS, 2018
Relevant Publications
- Alikhasi M., & Lelis, L. Unveiling Options with Neural Network Decomposition. In Proceedings of the International Conference on Learning Representations (ICLR), 2024.
- Carvalho, T.H., Tjhia, K., & Lelis, L. Reclaiming the Source of Programmatic Poli- cies: Programmatic versus Latent Spaces. In Proceedings of the International Conference on Learning Representations (ICLR), 2024.
- Moraes R., & Lelis, L. Searching for Programmatic Policies in Semantic Spaces. In the Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2024.
- Ameen S., & Lelis, L. Program Synthesis with Best-First Bottom-Up Search. Journal of Artificial Intelligence Research (JAIR), 2023.
- Aleixo D., & Lelis, L. Show Me the Way! Bilevel Search for Synthesizing Programmatic Strategies. In the Proceedings of the Conference on Artificial Intelligence (AAAI), 2023.
Orseau, L., Lelis, L., Lattimore, T., & Weber, T. (2018). Single-agent policy tree search with guarantees. In Advances in Neural Information Processing Systems (pp. 3201-3211).