Hit enter to search or ESC to close

Dr. Yisong Yue Thinks the Key To Breakthroughs in Autonomous Driving is All About Collaboration — and an Open Mind

Dr. Yisong Yue is pictured in front of the Los Angeles Skyline for Argo AI

“One of the things that I loved about being a student was there were just all of these different ideas,” says Yisong Yue, a professor of Computing and Mathematical Sciences at Caltech. “I loved hearing everyone’s viewpoints, and I continue to value that as a professor.”

In addition to having the unique ability to both give and receive knowledge, Yue is the kind of scientist who thinks that academics and the institutions they represent can create opportunities for students by forging partnerships with the industries they’re preparing students to enter. For Yue, that means taking a part-time sabbatical from Caltech to serve as a principal scientist at Argo AI’s newest engineering and development office in Los Angeles, California.

“We know we don’t have all of the solutions, we know that we need to do research, but we also want to be making tangible progress on improving the technology in a timely fashion,” says Yue, who will spend his time at Argo’s LA office sharing his expertise in machine learning to help the company continue its quest to design and build the safest, most effective self-driving technology. “These industry academia partnerships are created to accelerate that process.”

Yue’s first love is computer science. But he also has developed a passion for photography, choral music and basketball. In this confluence of technical and creative thinking, Yue discovered early on that developing both style and mechanics are the things he enjoys most in his work. One could deduce, then, that in Yue’s eyes there’s a relationship between science and style in every area of interest.

“If you can figure out the right way of understanding something, the answer is rather simple, artistic and elegant,” says Yue. “I appreciate elegance.”

Yue started thinking about technology this way during his childhood, when science fiction, video games and reading all piqued his interest. “I liked the fact that you had to be very logical, because the computer does exactly what you tell it to—no more, no less,” he says.

He went on to attend the University of Illinois Urbana-Champaign, where he got his bachelors in computer science before getting his doctorate in computer science from Cornell University. Through learning and practicing programming techniques like writing software, Yue has also grown to understand the importance of research—a belief that he still holds to this day.

It’s been over 15 years since he started studying machine learning and almost eight since he started teaching at Caltech. Now, he’ll bring his experience to Argo’s newest office.

Yue describes machine learning as the process of converting data and experience into knowledge. He points to the agricultural, industrial, electrical, and digital revolutions as instances in history where new technologies transformed history, as well as to the new era of automation we find ourselves in today.

“The machine learning—or AI—revolution says, now that we’re collecting a lot of data, a machine can look at that data and figure out how to do it automatically,” he says. “Or empower the human to do it more efficiently.”

When asked what hurdles he expects as he begins his dive into the machine learning research process at the Argo Los Angeles office, Yue shared that it all goes back to data, and it comes in stages: There’s the perception stack, the forecasting—then the planning.

He explains that humans, when driving, consider numerous possibilities of how to get from point A to point B before making a decision. That’s the planning part of it—a very cerebral, human function that you now need a computer to do, too.

“Autonomous vehicles use a combination of sensors, and the measurements from those sensors then have to be converted into objects,” Yue says. “For example, I can perceive these things in the world—this one looks like a pedestrian whereas that one looks like another car.”

From there, one of the most exciting elements of machine learning takes place: the machine learns to automatically extract rules based on the data. Forecasting, Yue says, means now that I have the data, I need to figure out what to do next. It asks, “can I predict what’s going to happen in the first 10 seconds after and then what about within the next ten seconds?”

In working on forecasting and planning with autonomous vehicles at Argo, Yue says that they’ll need to be able to understand uncertainty, however high- or low-stakes it may be. Then, with all of the possibilities to consider, they’ll need to focus on how to balance a multitude of criteria with making appropriate decisions on the fly.

“To build software that works on hardware like self-driving cars,” Yue says, “everything has to work: the perception, the forecasting, the planning, and the hardware that you design to support the software has to be reliable. This requires people with a range of perspectives and from different fields to work together.”

It requires, then, a person just like Yue.

One aspect of Yue’s career that has been very important to him is open access to scientific research. He is a strong proponent of Green Open Access, a category for scholarly publications and research where authors don’t need to pay to publish, and readers don’t need to pay to read. Yue feels that paywalls are a detriment to research, especially since we’ve transitioned from the print world to the digital world.

“The computer science field grew up in the computer age,” he says. “So, we have long prioritized openly sharing information on the internet. Like many computer scientists, I strongly support Green Open Access and I urge my colleagues in other fields to make that transition.”

It’s a transition, Yue admits, that will not be easy, as some of the world’s most prestigious scientific journals remain behind paywalls. Without the paywalls of the past, he contends, access to research is not determined by how much money someone has or doesn’t have—a fact that he says is integral to attracting traditionally underrepresented students within the realm of computer science.

At Caltech, Yue remained committed to paying it forward through one-on-one mentorships, fostering inclusive research communities, and fighting for open access to information. Now, he’ll be applying his unique brand of scientific and stylistic inclusivity at Argo, to push AVs that much closer to reality.

Learn more about Argo’s newest engineering office in Los Angeles here.

Choose your lane

How Autonomous Vehicles Distinguish Between Bicycles and People Who Ride Them

How Autonomous Vehicles Distinguish Between Bikes and People

When it comes to how autonomous vehicles see the world, humans come first, literally. Autonomous vehicles (AVs), like the kind operated by Pittsburgh-based Argo AI, use Machine Learning to detect and classify the objects in their surroundings, identifying people...
Why The League of American Bicyclists is optimistic about autonomous vehicles

Why a Leading Cycling Advocacy Group Is Optimistic About Autonomous Vehicles

As autonomous vehicle use grows, AV companies and the League of American Bicyclists are collaborating on how to ensure cyclists and motorists can share the roads safely, even if the “motorist” is artificial intelligence software. As part of the...

Self-Driving Is Arriving Right On Time. Just Like Ice Cream Did

Seven years ago, I was a self-driving skeptic. Not of the technology. Of all the “experts” promising autonomous vehicles would be everywhere by 2020. You didn’t need to be Nostradamus to know that was ridiculous. All you needed was...
Illustration of a futuristic parking deck turned into a mixed-use space, with AVs driving by

How Autonomous Vehicles Could Help Transform Parking Lots

Researchers say it’s likely that autonomous vehicles (AVs) can help reduce the need for parking lots, opening more room for grass and trees and other elements of nature. It may not seem like it when you’re circling the block...
An illustration of an Argo autonomous vehicle in teal against a backdrop of buildings, a bicyclist, and research papers

7 Big Breakthroughs From Argo Research at CVPR 2022

The 2022 Conference on Computer Vision and Pattern Recognition (CVPR 2022) is nearly here. Thousands of computer scientists, software engineers, and researchers from around the globe will gather in New Orleans to review and discuss their latest work in...

Researchers Predict the Future With Lidar Data

Researchers at Carnegie Mellon University’s Argo AI Center for Autonomous Vehicle Research, a private-public partnership funded by Argo for advancing the autonomous-vehicle (AV) field, say they have come up with a way to use lidar data to visualize not...

Must Reads