Post-Graduate Stories

Guneet Singh Dhillon Will Develop New Methods Using “E-Values for Data Science and Risk Management”

As a 2026 CANSSI Distinguished Postdoctoral Fellow, Guneet Singh Dhillon will take part in a comprehensive program that involves teaching, interdisciplinary or applied collaboration, professional development, and a research project focusing on the development of new tools and methods using e-values as an alternative to p-values for data science and risk management applications. Guneet will work under the supervision of Professors Ruodu Wang (University of Waterloo) and Christopher Blier-Wong (University of Toronto).

Program: CANSSI Distinguished Postdoctoral Fellowships
Region: National
Date: 2026–2028

Project Focus Areas

Guneet’s work will focus on developing new tools for data science and risk management using e-values, a modern alternative to p-values with strong theoretical guarantees.

Traditional approaches often rely on unrealistic assumptions and cannot handle continuous monitoring or complex dependence. E-values can provide valid and safe inferences in these settings. Guneet will make significant advances towards e-value methods for data science and risk management applications, such as multiple testing of risk measures and sequential testing for modern machine learning systems, promising substantial impact on data science and risk management frameworks.

He will engage in immersive applied collaborations that connect statistical innovation with real-world problems and will interact with leading researchers in statistics and machine learning, particularly the growing e-value community. Guneet will also engage to advise on implementing the developed tools.

In addition to conducting collaborative research, Guneet will teach statistics courses, participate in development workshops, and gain professional experience by supervising students and presenting at conferences.

Getting to Know Guneet

Guneet Singh Dhillon completed his DPhil in Statistics at the University of Oxford (2021–2026), where his research focused on the robustness of modern predictive and generative machine learning systems. He was supervised by Professors Yee Whye Teh, Tom Rainforth, George Deligiannidis, and Arnaud Doucet, and was also awarded the Clarendon Fund Scholarship.

Before his DPhil, Guneet spent three years as an Applied Scientist at Amazon Web Services (2018–2021). He received a BSc in Computer Science (Honors and Turing Scholars Honors) and a BSc in Mathematics (Honors) from the University of Texas at Austin (2014–2018), where he also completed a certificate in Applied Statistical Modeling, a minor in Economics, and an honors thesis supervised by Professor Adam Klivans.

Guneet aspires to “become an independent researcher in the statistical sciences, with an emphasis on practical theory, that is, developing meaningful theoretical work with tangible real-world impact.” Through the CANSSI Distinguished Postdoctoral Fellowship, he aims to “engage in interdisciplinary collaborations that bridge statistics with other scientific fields” and to take the opportunity to “teach and mentor the next generation of researchers.”

The CANSSI Distinguished Postdoctoral Fellowship places strong emphasis on interdisciplinary projects with applied components, aligning closely with my research ambitions.