Impressions from the Computational Advertising Summer Program

working group in action at SAMSI

One of three working groups meeting during the second week of the summer program.

We asked several of the participants of the Computational Advertising summer program to give us their impressions of the program. Here are a few of them:

David Banks, Professor of Statistics, Duke University and one of the organizers of the program

“SAMSI summer programs are great.  I was previously involved with an outstanding one on psychometrics, and this new one on computational advertising is even better.  We have three working groups, and each team is writing a research paper.  One group has found that keyword bidding generally conforms to theoretically optimal behavior in a first-price auction, with some interpretable exceptions.  Another group is modifying the regularization analysis that was used to win the Netflix prize to improve a music recommendation system.  And the third group is analyzing co-cluster structure and time dynamics in click-through rates for on-line news articles.

“The best thing about the program is the energy within the working groups.  Each consists of a team of six to eight people, with widely different ages, perspectives, backgrounds, and technical skills. I’m sitting with the bidding behavior group now, and eavesdropping on my colleagues as pairs and triplets of the participants sort out anomalies, compare different visualizations, speculate about causal mechanisms, and debate competitive strategies. There is a real team feeling here, and I don’t get to experience that in my usual academic life.  It is a lot of fun.”

David Banks and Noah Silverman

Noah Silverman, UCLA (left) and David Banks, Duke (right) collaborating in a working group.

Reginald Roberts, Chief Statistician, Rapp

“David was awesome. What a great facilitator! I come from industry and it was intimidating being around so many academics. David made tremendous efforts to make sure our thoughts and opinions were integrated into the various streams of intellectual thought and that we were important piece in research teams in the second week.”

Tuwaner Lamar, Assistant Professor of Mathematics, Morehouse College

“I like the workshop portion of the conference.  This is good because it requires a deeper understanding to the topic.  A month ago I asked the question why this resurgence in this area of statistics?  Because of the research done for the problem I’m working on, with my group, I have the answer.  I, also, like the SAMSI healthy, nutritious unique dishes served at lunch and I value the discussions over lunch.”

Tuwaner Lamar

Reginald Roberts, Rapp

Piaomu Liu, Ph.D. candidate, University of South Carolina

“The two-week long SAMSI workshop experience probably has been the best part of my summer! Cutting-edge research presentations from both academia and the industry gave me, a graduate student in statistics exposure to different research areas. I am definitely continuing to learn more about computational advertising beyond the workshop. Whether you are new to computational advertising like me, or have begun researching in this area, the workshop is devoted to facilitating learning, discussions, and making connections.  David and Karem have the best smiles in the world!:) ”

Chirag Lakhani, Postdoctoral Research Associate, Electrical Engineering, North Carolina State University

“The computational advertising workshop has been a very fun and informative experience.  On the first day I came into the workshop half-asleep and not sure what to expect but I was greeted by a very animated and congenial statistics professor.  He was able to wake me up from my slumber and made me realize this would be a great experience.  It was great to meet so many people from both academia and industry.  I have learned a great deal about the inner workings of people in industry.  The problems we have worked on in the working group have also been very helpful.  It gave me a great opportunity to get my hands dirty with data and also learn great techniques in machine learning.  This may be one of the most informative workshops I have attended.  It is a great model for other workshops.”

Chirag Lakhani in a working group

Chirag Lakhani, NCSU

Yi Jiang, Rapp

“David gave us great experience with a computational advertising workshop that had both theory from academia and application from industry. He made tremendous efforts to arrange the presentations and projects so that we can practice what we just learned right away, particularly with the extra helpful guidance from him. He was awesome!!”

Zainab Jamal, HP

“I had a great experience at the conference and workshop. I found the talks at the conference where a great mix of industry perspectives and the academics perspectives. I especially liked the talks by Mark Handcock and Tian Zheng on social networks, Liang Zhang’s talk on personalization of content and Mike Taddy’s talk on  extracting information from unstructured text data and the talks by Christian Posse(linked in) and Neel Sudarsan (eBay).

“I found the workshop very useful as I learned to wrap my head around large data –  – from loading it to manipulating it to running models on it efficiently. The interactions with other participants was great as I felt one with the ‘geeks’ and learned a lot.

“And last but not least, Prof Banks was warm and helpful and full of great ideas to make us think and I really appreciated that.”

Giri Gopalan, Ph.D. Candidate, Statistics, Harvard University

“It has been a great privilege to attend the 2012 SAMSI Workshop on Computational Advertising. What has been particularly marvelous about the workshop has been the synergy fostered between academia and industry. The speakers come from a wide range of backgrounds including research labs, start-ups, and universities. This confluence of variegated individuals has supplied me with a unique perspective on  both the theory and application of ubiquitous modern topics such as high dimensional statistics, machine learning, and networks. This synergy has also been evident in the working groups. Being able to interact with individuals from academia, research labs, and universities has imparted an educational experience that I cannot imagine receiving anywhere else.”

Xiaoming Huo, Professor, IySE, Georgia Tech

“The talks in the first week were informative. The data sets that are presented on the last Friday are quite interesting. During the working group (2nd week), we seem to get a lot of things done. Overall, it is a good experience: we learned quite a few things; made some friends; and maybe more… ”

group in front of a restaurant

Some folks got together for dinner during the summer program.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s