Professor Ling Ren Honored with 2023 Google Research Scholar Award

Computer Science Professor Ling Ren researches ways to use privacy-preserving algorithms to help users conduct private searches online. This summer, he received a 2023 Google Scholar Award recognizing him for his efforts. 

Ren’s winning proposal, titled “Practical Single-Server Private Information Retrieval,” examines how to use crypto algorithms to enable searchers to maintain an essential level of privacy when conducting private searches on public databases. 

Currently, when a user conducts a query through a server, the server sees what the user is requesting before directing the user to the appropriate page. His work examines ways to provide more privacy for the user so that the server will not be able to identify specifics about the user’s search.

As Ren explains, “We want the server to help you and to respond to your request without knowing what you requested, without knowing what it did for you.”

To illustrate this concept, he uses the example of an individual conducting a private search on a public database in the medical domain:

Let’s say that there is a public database about all sorts of diseases. And you conduct a search within the database for Parkinson’s disease. The fact that you are searching for Parkinson’s disease probably means either you or your close family or friend has Parkinson’s or has a high risk for that disease. If we can enable private lookups, all the server will know from your activity is that this particular user made a query to the health database. They won’t know what particular topic the user is interested in.

Through his research, Ren hopes to ultimately expand the work that the server does without impacting the user’s experience.  

In the end, it will still send back a response to the user without knowing what that response actually is. All the server will know is that the user fetched some pages on that site as opposed to which exact pages the user requested. But, when the user receives the page, from the user’s perspective, it’s just as if it received the information in the normal way.

One of the biggest challenges, he notes, is that implementing these types of privacy protections is very time-consuming and expensive. For decades, researchers have been striving to find better and more efficient ways to implement these protections.

With conventional private information retrieval algorithms, the server will have to do quite a lot more work than it would have done without privacy. Essentially, the server is doing computation on encrypted data. What the server used to be able to do in milliseconds would turn into hours of work under these privacy protection algorithms. Our project aims to drive down those costs back to milliseconds. 

Ren sees his current research as fundamental to protecting privacy in daily online search activity. However, he cautions, that there is still a way to go between research and deployment. “The cost right now is still high. I think we need multiple iterations to get it to a truly usable form.”

Professor Ren would like to recognize his PhD student Muhammad Haris Mughees for his technical contributions to the research. 

The Google Research Scholar Program “supports professors who are pursuing research in fields relevant to Google.” For his winning proposal, Professor Ren will receive a one-time research gift of $60,000. 

He feels honored to be recognized by Google for his work on user privacy.  He’s heartened that Google is willing to invest in ways to improve personal security, even though it may be costly. 

“Their recognition reflects the shared value that we have to treat privacy with more importance. It’s very encouraging for me and for the field that Google is investing resources in privacy.”