October 17, 2013
Interview with CS Celebrity: Ben Zhao
Ben Zhao is a computer science professor at UCSB who made the TR35 list in 2006 and has since become a sought-after expert judge for innovator lists, hackathons and other competitions
by Sonia Fernandez
If it’s in the realm of social networks, Internet security and privacy, and the modeling and mining of enormous graphs, you can bet Ben Zhao is on the cusp of something big. An associate professor of computer science at UC Santa Barbara with degrees from Yale and UC Berkeley, Zhao put himself on the radar with recognition for his work, including an early career award from the National Science Foundation and a place on ComputerWorld’s list of Top 40 Innovators Under 40.
Zhao’s work on large scale peer-to peer-networks, a phenomenon that has exploded with the growth of the Internet, landed him on the list of MIT Technology Review 35 (Top Innovators under 35) in 2006, and made him the go-to guy for computer science topics, from hacking (a field in which he has served as a celebrity judge in ethical hacking competitions), to the organization of information in massive data sets, such as the ones generated by social networks like Facebook and Twitter.
We caught up with the perpetually busy Zhao for a brief discussion on his recent experience judging the 2013 TR35, a competition that he has been asked to judge ever since he earned a place on it years ago.
What was it like to judge the TR35? Out of the multitude of no doubt excellent entries you received, how did you manage to pick the winners?
It's fun, and I take it fairly seriously. It is a prestigious award, and any decision like this can potentially have a significant impact on the careers of young scientists. So I weigh the decisions carefully.
The process is fairly simple; each of the judges receives a list of roughly 10 profiles of candidates, each with multiple documents (their own writeup, a nomination letter, and multiple letters of support, with potentially other external documents like publications, etc). We each rate the candidates we see from a scale of 1 to 10, and write a review justifying our decisions/scores. Once all the results are tallied, the MIT Tech Review staff makes the final decisions.
Personally, I am looking for: A) clear impact in their field of work, B) a measure of the pure novelty or creativity of the idea or work done by the candidate, and C) the likelihood of long term productivity, i.e. is this a one hit wonder or the first of many innovations to come. It may be surprising to some, but it is actually quite easy to separate out the best candidates using these criteria. There is often a significant gap between Yes and No, and it is easy to tell quickly if a TR-35 candidate "has it."
Which ones were your personal favorites, and why?
Given my areas of expertise, the candidates I review tend to be in Internet/software areas. I didn't see an Elon Musk in my group, but there were several exciting candidates whose work signaled significant changes in the way we thought about or conducted research. I won't tell you who I reviewed, but looking at the entire set of TR35 winners this year, I am very excited by several folks who are using innovative applications of crowdsourcing to change the world. Anthony Goldbloom, the founder of Kaggle, is one. Rebeca Hwang has another with YouNoodle. There are always winners who look at how to dramatic change life in developing regions, especially with the use of mobile phones, e.g. Caroline Buckee, Bright Simons.
In you estimation, which kinds of projects are likely to make the cut? Are there particularly pressing problems that need solutions, or questions that need answering? What approaches work best, or have worked best?
Typically, the winners are working on the most important problems of our day, whether it's improving life in developing regions, or tackling big data, or dealing with tough diseases, or the rise of cybersecurity threats (Dmitri Alperovitch with Crowdstrike). Working on the right problems means a great deal. There are generally no real consistent methodologies on how to succeed, since a big part of the award is for creativity and out of the box ideas. But typically speaking, the solutions tend to be cheap, practical, and often involve the use of existing technology in unexpected contexts.
Ben Y. Zhao is an associate professor in the Department of Computer Science, part of the College of Engineering at UC Santa Barbara. Together with Prof. Heather Zheng, Zhao co-directs the SAND Lab (Systems, Algorithms, Networking and Data) at UCSB. He collaborates with other research groups in the department, including the Security Lab, the Database group, the Mobility Management and Networking (MOMENT) lab, and groups in the Communications, Electrical & Computer Engineering, and Statistics departments at UCSB.