When the Milwaukee School of Engineering opened its new $34 million Dwight and Dian Diercks Computational Science Hall in the fall of 2019, it was intended to not only provide students with hands-on learning opportunities, but to also foster collaboration with area companies.
Both for businesses that already rely on data science to inform their business decisions and those that are newer to exploring its potential, the artificial intelligence hall – equipped with a Nvidia GPU-accelerated supercomputer – invites companies to take advantage of its computing power and student manpower.
“The massive amount of resources that MSOE is bringing to bear in these areas, it has generated a tremendous amount of interest from a wide variety of constituents,” said Steve Williams, professor and chairperson of MSOE’s Electrical Engineering and Computer Science Department. “And I’ve observed that in industry and folks who are monetizing data science, there’s a great deal of variation in the maturity of those organizations. We have established fruitful working relationships with organizations that already have in-house talent that they’re looking to augment with our talent and some of our hardware resources.”
MSOE’s AI effort is one of several recently-forged partnerships between higher education and industry aimed at advancing data science in the state.
Last year, the American Family Insurance Data Science Institute opened on the University of Wisconsin-Madison campus as part of a $20 million commitment from the Madison-based insurance company. The investment included $10 million for research over the next decade and a $10 million endowment to establish the institute at UW’s McArdle Building, with the goal of accelerating studies in various fields, including artificial intelligence, genetics, drug development, material science and business.
Also last year, the Northwestern Mutual Data Science Institute – a nearly $40 million collaboration of the Milwaukee-based life insurance company, the University of Wisconsin-Milwaukee and Marquette University – opened on the company’s downtown Milwaukee headquarters. Since launching, the institute has initiated its first three research projects. They are focused on leveraging data science to address inequities in a Milwaukee neighborhood, the opioid crisis and 2020 voter sentiment.
From offering new insight into societal issues to optimizing business operations, companies are recognizing the competitive advantage of using data science to make decisions, solve problems and create new products. In response, higher education and industry are partnering to prepare a pipeline of workers who are trained to mine for meaning in massive data sets.[caption id="attachment_506120" align="alignright" width="300"] Glenn Fung[/caption]
In contrast to other science fields, data science has largely evolved out of those partnerships, said Glenn Fung, chief research scientist, AI and machine learning research director at American Family.
“The technology is showing so much promise, so much value in applications,” he said. “It’s one of the first times in the history of science, I think, where there is a modality of science that has been driven by companies and especially by universities.”
While the insurance industry isn’t necessarily associated with pushing the envelope on innovation, it has relied on data-driven decisions for a long time, he said.
“Traditional insurance, it’s been thought of as very conservative,” said Fung, who worked for Siemens and Amazon before joining American Family. “... But insurance actually was one of the first … kinds of models of business that has used data from the beginning … We in insurance had to understand what was the frequency of accidents in order to see what was the best way to provide coverage … It’s actually the foundation of data science.”
But, to remain competitive, it’s important for the company to stay on the forefront of data-driven innovation, Fung said.
Before committing $20 million to UW last year, American Family had invested in multiple research projects at the university. Under the new partnership, the company is now solidifying that partnership by awarding $1 million in research grants annually.
“We wanted to get a process because, before, … I knew professors, I knew the cool things that they were doing and I talked to them and we came into these research partnership agreements and they were resulting in interesting stuff for us and also for the professors, but it was kind of an informal process,” Fung said.
The institute recently selected its first round of 10 research projects from a pool of 37 proposals across the university.
One of the projects directly applies to American Family’s work. UW-Madison experts within the Department of Atmospheric and Oceanic Sciences and the UW Space Science and Engineering Center are creating wind-related weather models to more accurately predict natural events like hailstorms, hurricanes, wildfires and tornadoes.
These models could help the company prepare for insurance claims and more quickly mobilize resources and claims adjusters to serve customers, Fung said.
MSOE sees its AI initiative as part of a larger ecosystem that’s developing in the region.
“IT (information technology) moved us in the direction of a knowledge-based economy and this is the next phase of evolution of that,” Williams said. “And as we start graduating students with the skill sets … we’re going to see a feedback loop where more and more companies will have some of this talent and they’ll come back to us and we’ll work on more interesting projects that will end up making them more money and the economy will grow from that.”
A core component of MSOE’s computer science program is offering students real-world projects with industry partners. So far, the program has attracted interest from area companies, including those that are just dipping their toes in data science.
“Like any burgeoning industry, there are some companies that have been doing this for a long time and have a plan together and are really interested in the emerging fronts that exist in this space because it’s a very fast-moving area,” said Derek Riley, program director of MSOE’s computer science program. “... And then there are a lot of companies that realize there’s an opportunity here; it’s not uncommon that a company says, ‘Hey, we have a data set. We know there’s potential there. Can your students work on a project?’”
Sometimes, it’s a good fit, and other times, a business’s pitch isn’t conducive to a student project, Riley said. One issue, he said, is the company might not have the infrastructure in place to fully support the project.
“There are a lot of organizations that have the data, they have a problem they really want to solve with the data, but even if we delivered a solution to them, these things are not static systems; they have to be updated and maintained,” Riley said.
Among industry partners that have approached MSOE, Williams said health care organizations tend to be ahead of the manufacturing sector in taking advantage of data.
“If you’re looking at a manufacturing process, where, to put these tools in place, they have to have this huge investment in better sensing capacity and better processing, the investment can be a pretty big barrier to the automation or AI implementation,” he said. “There are always ways to improve things, it’s just that cost-benefit factor that ends up being a barrier for a lot of (companies).”
He added that some manufacturers are also playing catch up to the health care and other industries when it comes to talent development.
Providing talent for industry is MSOE’s goal. In less than two years since MSOE launched its computer science degree, it has become the second-highest-enrolled degree program at the college.
“Demand for the skill set is understood by the students who are choosing majors,” Riley said. “It’s pretty clear that AI, data science, it’s an in-demand field that’s not going away.”
And for students who want to stay in the area after graduation, there are plenty of career opportunities in data science locally, Riley said.
“Milwaukee and Madison together, they have a huge number of companies looking for people with these skill sets,” he said. “There are really large companies hiring these folks, companies like Northwestern Mutual and Kohl’s, Milwaukee Tool … Epic, Rockwell Automation, but there’s also a surprising number of startup companies, consulting companies and other smaller companies that are getting into this space and have a large demand.”
While COVID-19 has changed the workforce landscape, Riley said he sees companies continuing to invest in future workers.
“Companies are hiring interns as a feeder recruitment pipeline,” Riley said. “They give the students really interesting and exciting problems to try to sell themselves as a future employer. Students are working on really interesting stuff as this kind of recruitment strategy. I think companies are keeping that alive because they see the demand for the skill set is so strong, if they don’t go out there and compete for students now, they’ll miss out.” ϖ