From schneide at cs.cmu.edu  Thu Feb 26 13:05:58 2015
From: schneide at cs.cmu.edu (Jeff Schneider)
Date: Thu, 26 Feb 2015 13:05:58 -0500
Subject: **REMINDER** ML/Google Distinguished Lecture - Peter Frazier -
 February 26, 2015
In-Reply-To: <081339BF5B085646911B69780926302C0686358846CC@EXCH-MB-1.srv.cs.cmu.edu>
References: <081339BF5B085646911B69780926302C0686358846CC@EXCH-MB-1.srv.cs.cmu.edu>
Message-ID: <54EF6086.3050101@cs.cmu.edu>


This should be a good talk.  Hope to see you there!
Jeff.



-------- Forwarded Message --------
Subject: 	**REMINDER** ML/Google Distinguished Lecture - Peter Frazier -
February 26, 2015
Date: 	Wed, 25 Feb 2015 14:26:49 -0500
From: 	Sandy Winkler <sandyw at cs.cmu.edu>
To: 	ml-seminar at cs.cmu.edu <ml-seminar at cs.cmu.edu>
CC: 	sandyw+ at cs.cmu.edu <sandyw+ at cs.cmu.edu>



Please join us for the ML/Google Distinguished Lecture!

*Thursday, February 26, 2015*

*4:30 pm*

*NSH  3305*

_Peter Frazier,
<http://stat.cornell.edu/people/field-faculty/peter-frazier>_Assistant Professor
Operations Research and Information Engineering (ORIE)
<http://stat.cornell.edu/department/operations-research-and-information-engineering-orie>, 

Cornell University

Host: Jeff Schneider

Title: Parallel Bayesian Global Optimization of Expensive Functions, for Metrics
Optimization at Yelp

Abstract:

We consider parallel derivative-free global optimization of
expensive-to-evaluate functions. We present a new decision-theoretic algorithm
for this problem, which places a Bayesian prior distribution on the objective
function, and chooses the set of points to evaluate next that provide the
largest value of the information.  This decision-theoretic approach was
previously proposed by Ginsbourger and co-authors in 2008, but was deemed too
difficult to actually implement in practice. Using stochastic approximation, we
provide a practical algorithm implementing this approach, and demonstrate that
it provides a significant speedup over the single-threaded expected improvement
algorithm. We then describe how Yelp, the online business review company, uses
this algorithm to optimize the content that their users see.  An open source
implementation, called the Metrics Optimization Engine (MOE), was co-developed
with engineers at Yelp and is available at github.com/yelp/MOE
<http://github.com/yelp/MOE>.

Bio

Peter I. Frazier is an assistant professor in the School of Operations Research
and Information Engineering at Cornell University, and received a Ph.D. in
Operations Research and Financial Engineering from Princeton University in
2009.  He is the recipient of an AFOSR Young Investigator Award, and an NSF
CAREER Award.  He is an associate editor for Operations Research, ACM
Transactions on Modeling and Computer Simulation and IIE Transactions.  His
research interest is in dynamic programming and Bayesian statistics, focusing on
the optimal acquisition of information and sequential design of experiments.  He
works on applications in simulation, optimization, operations management,
medicine, and materials science.

Sandra Winkler

*Machine Learning Department*

*Carnegie Mellon University*

5000 Forbes Avenue

Pittsburgh, PA  15213-3890

sandyw at cs.cmu.edu <mailto:sandyw at cs.cmu.edu>

Tel: 412-268-3569

Fax: 412-268-2205