From left to right: Carlos Cinelli, Armeen Taeb, Alexander Giessing, and Vincent Roulet

The Department of Statistics is pleased to welcome Carlos Cinelli, Armeen Taeb, and Alexander Giessing, and to congratulate Vincent Roulet in his new position!

Carlos Cinelli will be joining our department as an Assistant Professor on September 16, 2021. He will be graduating with his Ph.D. in Statistics from the University of California, Los Angeles in early-September 2021. 

Cinelli’s research focuses on developing new theory, methods, and software to help data scientists make reliable causal inferences under realistic settings – more specifically, many of these tools enable empirical scientists to assess the robustness of their scientific findings against plausible violations of traditional assumptions. He is particularly interested in the inferential challenges faced by social and health scientists, as well as the intersection of causality with machine learning and artificial intelligence. His work has resulted in publications in the most prestigious outlets of statistics, epidemiology and machine learning. Most notably, these results are having a real impact in research practice and have been applied by many scientists across different disciplines, ranging from economics, political science, neurodevelopment, epidemiology and genetics.

Cinelli expressed with enthusiasm, “I’m super excited about joining UW, especially due to its strong interdisciplinarity. Two of my current projects consist of: (i) continuing the development of theory and software to make sensitivity analysis easy to perform routine and standard practice across the empirical sciences; and, (ii) developing the foundations for a more flexible approach to causal inference with credible assumptions. I believe that initiatives such as the Center for Statistics and the Social Sciences, as well as the strong connections of Statistics with Biostatistics and Computer Science will provide an invaluable environment towards achieving this goal. I am also very enthusiastic about contributing to the methodological training and mentoring of students.”

Armeen Taeb will be joining our department as an Assistant Professor on September 16, 2022. He completed his Ph.D. in Electrical Engineering at California Institute of Technology under the supervision of Professor Venkat Chandrasekaran in 2019. He is currently a Postdoctoral Associate of ETH Foundations of Data Science (ETH-FDS) at ETH Zürich.

Taeb’s research interests involve the intersection of statistics and optimization. His work focuses on developing methods for latent-variable modeling as well as techniques to control false discoveries in contemporary data analysis settings. He is also interested in exploring the utility of statistical methodologies for real-world applications such as water resource management. During his Ph.D. thesis, Taeb received the W.P. Carey & Co. Prize for outstanding dissertation in Applied Mathematics for his paper on “Latent-variable modeling: inference, algorithms, and applications”. In addition, because of his research on developing statistical models for the California reservoir system, he was awarded the Resnick Institute Fellowship for Sustainability Research.

“I look forward to joining the Department of Statistics and engaging with like-minded researchers who are focused on developing principled methodologies with real impact,” says Taeb. “I am also excited to interact with and learn from the excellent students in the department! Finally, I look forward to enjoying the beautiful nature in Seattle and going on many nice hikes.”

Alexander Giessing will be joining our department as an Acting Assistant Professor on September 16, 2021. He received his Ph.D. in Statistics from the University of Michigan, Ann Arbor, under the supervision of Professor Xuming He in 2018. He is currently a Postdoctoral Research Associate at Princeton University under the supervision of Professor Jianqing Fan.

Giessing focuses on theory and methodology of robust statistics for large, high-dimensional, and highly granular data. In recent years, the ready availability of such data has spurred exciting innovations in precision medicine and personalized consumer services. However, high granularity is both a boon and a bane and usually comes at the cost of increased heterogeneity, additional variability, heavy tails, and missingness among other things. In his research, he develops statistical procedures to address these issues. Currently, he is working on (i) the theoretical foundations of bootstrap for high-dimensional data, (ii) bootstrap-tests for multiple and global hypothesis testing, and (iii) methodology for high-dimensional quantile regression with applications to treatment effect estimation.

In the upcoming year, Giessing looks forward to carry out his research agenda on bootstrap procedures for high-dimensional data. “Bootstrap methods are extremely versatile, but their scope of applicability is far from fully understood in modern, high-dimensional statistical problems,” says Giessing, “I look forward to collaborating within and beyond the department on possibly more data-driven problems in robust statistics. I find that real data sets are still the best source of inspiration for novel statistical methodology.”

During his free time, Giessing likes to go on runs in the mornings and improvise on the piano in the evenings. Now that he is in Seattle, he looks forward to adding variety with an occasional hike or skiing trip on the weekends.

Vincent Roulet is currently an Acting Instructor in our department. He will begin his new position as an Acting Assistant Professor on September 16, 2021. Roulet has been with the Department of Statistics since March 2018. He first joined our department as a Postdoctoral Scholar, under the supervision of Associate Professor Zaid Harchaoui. He then transitioned to an Acting Instructor position in September 2020. 

Roulet completed his Ph.D. in Applied Mathematics (with a focus on Optimization and Machine Learning) from Paris Sciences et Lettres University, under the supervision of Professor Alexandre d’Aspremont in 2017. His dissertation demonstrated both theoretically and empirically how accelerated algorithms for convex optimization can be sped up by restarts. These results led to refining complexity measures, such as sparse recovery problems. He is currently working on the optimization of non-linear dynamical problems arising, for example, in non-linear control or deep learning. He is studying global convergence guarantees and efficient algorithms using the structure of the problems. 

“I’m looking forward to collaborating with students and faculty, to bring my optimization knowledge and learn more statistical models and applications,” says Roulet. “I also look forward to hiking more in the beautiful surroundings of Seattle in the next years!”