I'm interested in modeling complex real-world phenomena with simple equations and reducing them to problems that can be solved with practical computational and memory resources. Darwin elegantly explained the diversity of life with the theory of evolution driven by natural selection. Larry Page and Sergey Brin calculated an importance of web pages by a simple matrix operation called PageRank. These are examples of Occam's Razor: that it is not necessarily the most abstruse and complex theories that are the most impressive, but those that are simpler than the structure of the phenomenon or data might suggest. In today's world where the use of big data has become routine, I’m paradoxically seeking mathematical models that do not depend on high-performance computing and massive data sets, yet can still extract essential information from data.

[Keywords]
Machine learning / Data analytics / Statistical learning / Causal inference / Reinforcement learning