There’s online teaching and then there’s cool YouTube-style teaching with a lightboard, neon-colored markers, a black backdrop and a Steve Jobs-worthy wardrobe on the instructor.
Steve Brunton is doing the latter.
A mechanical engineering professor at the University of Washington, Brunton records videos from a small studio on campus. The videos have titles such as “Neural Network Architectures and Deep Learning”; “Linear Quadratic Regulator (LQR) Control for the Inverted Pendulum on a Cart”; “Underdetermined Systems and Compressed Sensing” and many more. His YouTube channel has attracted more than 95,000 subscribers and four million views.
“My work has always straddled engineering and applied math,” said Brunton, our latest Geek of the Week. “I love teaching, especially math to engineers, and a lot of my effort goes into making educational videos and textbooks for a broad audience.”
Brunton goes by “Eigensteve” — a reference to the linear algebra terms eigenvalue and eigenvector — and in a profile on the UW Mechanical Engineering news page, he said he’s been surprised by the overwhelmingly positive reaction to his channel, because “people tend to trash everything on the internet.”
According to UW, Brunton began recording lessons after joining the university as an acting assistant professor in applied mathematics. He and another professor were teaching AMATH301: Beginning Scientific Computing, one of engineering’s core classes taken by over a thousand students a year.
They decided to “flip” the class because of its repetitive nature and little change to the curriculum. This involves reversing the typical lecture-then-homework model. Instead, students watch lecture material online ahead of class, and class time is spent engaged in problem solving activities.
Recorded lessons were better at reaching more students, Brunton told UW, and now hundreds of videos later, with a newer studio and upgraded technical capabilities, it’s easy to scroll through the channel and see how the production value has improved over time.
“I usually get a few emails every week from instructors at other institutions asking for advice about the setup,” Brunton said. “A lot of people have pioneered flipped classrooms and asynchronous teaching using other technologies. But I think the lightboard studio is a really compelling format.”
When he’s not on YouTube helping to simplify the fundamentals of applied math, Brunton enjoys watching videos about bartending and cocktails (The Educated Barfly is his favorite), and cooking (Babish). His family was even inspired to make their own stop-action cocktail video at the beginning of the pandemic lockdown.
Brunton grew up in Texas and met his wife Bing at California Institute of Technology during pre-freshman orientation weekend. Bing Brunton is an associate professor of biology at UW and the couple has two kids and a dog.
Brunton has always had a passion for science fiction, especially space travel and future technology. And he shared these additional fun facts about himself: “I carve my own crystal clear ice for drinks; I wear one of 10 identical black shirts every day; I’ve never once had a professional haircut; and I can count to over 1,000 on my 10 fingers.”
Learn more about this week’s Geek of the Week, Steve Brunton:
What do you do, and why do you do it? I’m a professor at UW, so I wear a lot of hats. But at the end of the day, there are three things I work on that really excite me: 1) I teach math to engineers; 2) my lab works on fundamental research to marry machine learning with classical engineering problems in dynamical systems modeling and control; and 3) we apply this research to better understand and control systems in fluid dynamics.
- I teach math to engineers: I absolutely love teaching. It is by far the most fun and rewarding part of my job, having an impact on young people who are eager to improve themselves. My life has been defined in large part by the efforts of great teachers who have been generous with their investment in me, and I am fortunate to have the opportunity to invest in the next generation. I spend a lot of time (mostly nights and weekends) writing books and making YouTube videos, which has been very rewarding.
- Fundamental research in ML for dynamical systems and control: Ever since I learned about chaos in the planetary three-body problem, I have loved dynamical systems. Dynamics and control are really the culmination of two of the most useful fields of applied math: differential equations and linear algebra. These fields are currently undergoing a revolution, with data-driven optimization and regression techniques (i.e., machine learning) redefining what is possible. This is a very exciting time to be working in data-driven modeling and control.
- Applied fluids research: Humans have a deep connection to fluid dynamics. We literally spend our entire lives living in and breathing fluids, fluids pump through our veins, and most of our engineering devices also operate in a fluid environment. Besides being beautiful, fluids are critically important for many technologies. Many science fiction technologies of the future will require an improved ability to model and control fluids.
What do all of these have in common? Chaos is sometimes defined as a sensitive dependence on initial conditions, where small perturbations early can have huge downstream effects; this is the fundamental principle of flow control and also of educating young people.
What’s the single most important thing people should know about your field? Many tasks in classical engineering can be posed as an optimization problem (reduced-order modeling, optimal control, sensor placement, etc.). For modern systems of interest, such as turbulence, neuroscience, climate modeling, and epidemiology, these optimizations are extremely nonlinear, non-convex, and high-dimensional. Emerging methods in machine learning and artificial intelligence are essentially new optimization and regression techniques based on data that are well suited for these highly non-convex problems. So the classical and modern approaches almost seem tailor made for each other.
Where do you find your inspiration? I decided to go to grad school to study fluid dynamics after seeing the vortical patterns in a bowl of miso soup and realizing that it was more interesting than what I was working on.
But seriously, mostly in my students and collaborators, who help me blur the line between work and play.
What’s the one piece of technology you couldn’t live without, and why? My Bialetti Moka coffee pot.
What’s your workspace like, and why does it work for you? I try to work outside as much as possible, although it is harder in the winter.
Your best tip or trick for managing everyday work and life. (Help us out, we need it.) I try to take walking meetings whenever possible. I also prep dinner and drinks during my large blocks of Zoom meetings.
Mac, Windows or Linux? Mac.
Kirk, Picard, or Janeway? Captain Jean Luc Picard of the USS Enterprise (volume up).
Transporter, Time Machine or Cloak of Invisibility? Transporter. Travel + time.
If someone gave me $1 million to launch a startup, I would … develop a smart toilet for automated health data.
I once waited in line for … “The Matrix Revolutions” (so disappointing).
Your role models: Personally, definitely my parents and my wife. Professionally, I would say the great teachers that have inspired me throughout my life: My dad, John Quintanilla at UNT/TAMS, Jerry Marsden at Caltech, Nathan Kutz at UW; also the popular TV scientists like Julius Sumner Miller and Carl Sagan.
And my dog, who is a spiritual air purifier.
Greatest game in history: “Quest for Glory” (EGA).
Best gadget ever: My double pendulum fidget spinner.
First computer: Home built PC with Intel 486 running DOS
Current phone: iPhone 11.
Favorite app: Duolingo.
Favorite cause: Female entrepreneurship and general education in developing countries.
Most important technology of 2020: Zoom.
Most important technology of 2022: Robust supply chains.
Final words of advice for your fellow geeks: Stay curious!
Website: Brunton Lab
Twitter: @eigensteve
LinkedIn: Steve Brunton