I research causal inference and its pivotal role in establishing generalizable machine learning. In particular, I utilize the formal language of causality to understand the fundamental challenges of prediction in the out-of-distribution (OOD) generalization problem. My primary focus is on the theoretical aspects of the topic, such as the asymptotic properties of learning algorithms and their performance guarantees.
Before joining Columbia, I completed my undergraduate degree in Computer Science and Economics at Sharif University of Technology, Tehran, Iran.
Email: kasra at cs dot columbia dot edu
Transportable Representations for Out-of-distribution Generalization
K. Jalaldoust, E. Bareinboim.
Columbia CausalAI Laboratory, Technical Report, May, 2023.
Causal Discovery in Hawkes Processes by Minimum Description Length
K. Jalaldoust, Katerina Schindlerova, Claudia Plant
AAAI-2022. In Proceedings of the 36th AAAI Conference on Artificial Intelligence.
Oral Presentation (<1%, out of 9020 papers)