Peter I. Frazier
Associate Professor, Cornell University
School of Operations Research
and Information Engineering
Staff Data Scientist, Uber
My group works on optimal learning and the exploration vs. exploitation tradeoff,
at the interface between machine learning and sequential decision-making under uncertainty.
- [Jul'17] Starting July 2017, I am working at 50% effort for Uber as a Staff Data Scientist while based in Ithaca, and 50% for Cornell as an Associate Professor. This follows a 2-year leave at Uber where I managed data science for uberPOOL, Uber's carpooling product.
- [May'17] Congrats to postdoc Matthias Poloczek, who will be joining University of Arizona as an Assistant Professor.
- [Jan-Aug'17] Congrats to PhD students Jialei Wang, Steve Pallone, Weici Hu, and Jian Wu, who recently graduated from the group, and will be joining IBM, Uber, Google, and AQR.
- [Jan'16] Congrats to my former PhD student Scott Clark, just named one of
Forbes' 30 under 30 for his Bayesian Optimization startup, SigOpt!
In collaboration with Yelp, we've released
Metrics Optimization Engine (MOE), an open-source service for Bayesian optimal experimental design.
- For a general audience,
"Using machine machine learning and optimization to improve Yelp's website" (slides,
starting at 18:30 in this video).
- For PhD students and other researchers, see these slides from a two-day masterclass at Lancaster University:
- Other surveys, tutorials, and videos, may be found on the intro page.
- MIT, OR Center, Feb 2016.
Parallel Bayesian Global Optimization of Expensive Functions, for Metrics Optimization at Yelp
- Carnegie Mellon, Machine Learning Department, Feb 2015.
Optimal Learning for Molecular Discovery
- Cornell Center for Applied Math, Oct 2014.
Information Filtering for arXiv.org: Bandits, Exploration vs. Exploitation, and the Cold Start Problem
- Georgia Tech, School of Industrial and Systems Engineering, Apr 2014.
Ranking and Selection With Tight Bounds on Probability of Correct Selection
- Simulation Optimization Workshop, Vina del Mar, Chile, Mar 2013.
- For others, see the Presentations page. For a complete list, see my CV.
B. Chen, P.I. Frazier,
Dueling Bandits with Weak Regret,
Wu, J., Frazier
The Parallel Knowledge Gradient Method for Batch Bayesian Optimization,
Frazier, D. Kempe, J. Kleinberg, R. Kleinberg,
Economics and Computation, 2014.
A Fully Sequential Elimination Procedure for Indifference-Zone Ranking and Selection with Tight Bounds on Probability of Correct Selection,
Operations Research, 2014.
J. Xie & Frazier,
Sequential Bayes-Optimal Policies for Multiple Comparisons with a Control,
Operations Research, 2013.
Frazier, W.B. Powell & S. Dayanik,
A Knowledge-Gradient Policy for Sequential Information Collection,
SIAM Journal on Control and Optimization, 2008.
- For a full list see the
or my CV.