Assistant Professor of Electrical Engineering and Computer Science at UCI Samueli School of Engineering
Researcher Spotlight: Zhou Li
What brought you to UCI?
I was a principal research scientist at an industrial research laboratory called RSA Labs before coming to UCI, focusing on developing new technologies that can be transferred to company’s products. The main reason for moving from industry to university is that I can work on very challenging problems calling for very innovative solutions, different from industrial research which tends to be short-term and less risky. This makes me quite excited. Another reason is that I’ll have chance to teach and mentor students and see them advancing to the next level.
I chose UCI because it has very strong programs related to computing technologies and especially information security. The students and faculty are very talented and many great works have been done at UCI. And of course, Irvine is a wonderful place to live.
What is your major focus area as a researcher, and why?
My main focus is the intersection between data and security. I believe there are a lot of open problems to solve at this intersection. Now data are generated at an amazing pace by various computing devices and there is strong need to develop new data-driven methods that can solve challenging problems, like the ones in security.
In one sentence, what is the most important question you want to address?
The questions I’m trying to answer include: how to unlock the power from big data to catch the cyber-attackers so our computing systems could be more secure? what new threats are emerged because of the data generated by the new computing devices/hardware?
What has been (or will be) the impact of your research?
Some of my research at RSA Labs have been transferred into company’s products in security analytics. The techniques we built enabled the detection of very sophisticated hacking activities. Besides, my research has identified critical vulnerabilities underlying new computing devices/hardware like smartwatch, mobile payment and GPU, due to data leakage. We worked with the vendors to fix the vulnerabilities and make their devices more secure.
What is innovative about your research?
The key challenge I’m facing in most of the projects is how to make best use of data. To this end, I built many customized machine-learning models to address different security problems. As an example, one research during my time at RSA Labs combined supervised learning, temporal analysis and graph-based inference together to identify malicious domains visited by employees. It is able to examine tera-bytes of log data generated every day from enterprise devices and find most of malicious domains with very high accuracy.
What papers do you have coming through in the next year?
I have one paper accepted by NDSS’19 on security analysis of Bluetooth protocols and another one accepted by IEEE S&P’19 reveals the fraudulent behaviors of residential IP proxy so far.