Tamara Broderick first set foot on MIT’s campus when she was a highschool student, as a participant within the inaugural Women’s Technology Program. The monthlong summer academic experience gives young women a hands-on introduction to engineering and computer science.
What’s the probability that she would return to MIT years later, this time as a school member?
That’s an issue Broderick could probably answer quantitatively using Bayesian inference, a statistical approach to probability that tries to quantify uncertainty by repeatedly updating one’s assumptions as recent data are obtained.
In her lab at MIT, the newly tenured associate professor within the Department of Electrical Engineering and Computer Science (EECS) uses Bayesian inference to quantify uncertainty and measure the robustness of information evaluation techniques.
“I’ve all the time been really eager about understanding not only ‘What can we know from data evaluation,’ but ‘How well can we understand it?’” says Broderick, who can be a member of the Laboratory for Information and Decision Systems and the Institute for Data, Systems, and Society. “The fact is that we live in a loud world, and we will’t all the time get exactly the info that we wish. How can we learn from data but at the identical time recognize that there are limitations and deal appropriately with them?”
Broadly, her focus is on helping people understand the confines of the statistical tools available to them and, sometimes, working with them to craft higher tools for a selected situation.
As an illustration, her group recently collaborated with oceanographers to develop a machine-learning model that could make more accurate predictions about ocean currents. In one other project, she and others worked with degenerative disease specialists on a tool that helps severely motor-impaired individuals utilize a pc’s graphical user interface by manipulating a single switch.
A standard thread woven through her work is an emphasis on collaboration.
“Working in data evaluation, you get to hang around in everybody’s backyard, so to talk. You actually can’t get bored because you may all the time be learning about another field and fascinated about how we will apply machine learning there,” she says.
Hanging out in lots of academic “backyards” is particularly appealing to Broderick, who struggled even from a young age to narrow down her interests.
A math mindset
Growing up in a suburb of Cleveland, Ohio, Broderick had an interest in math for so long as she will be able to remember. She recalls being fascinated by the thought of what would occur for those who kept adding a number to itself, starting with 1+1=2 after which 2+2=4.
“I used to be possibly 5 years old, so I didn’t know what ‘powers of two’ were or anything like that. I used to be just really into math,” she says.
Her father recognized her interest in the topic and enrolled her in a Johns Hopkins program called the Center for Talented Youth, which gave Broderick the chance to take three-week summer classes on a variety of subjects, from astronomy to number theory to computer science.
Later, in highschool, she conducted astrophysics research with a postdoc at Case Western University. In the summertime of 2002, she spent 4 weeks at MIT as a member of the first-class of the Women’s Technology Program.
She especially enjoyed the liberty offered by this system, and its give attention to using intuition and ingenuity to attain high-level goals. As an illustration, the cohort was tasked with constructing a tool with LEGOs that they may use to biopsy a grape suspended in Jell-O.
This system showed her how much creativity is involved in engineering and computer science, and piqued her interest in pursuing a tutorial profession.
“But once I got into college at Princeton, I couldn’t determine — math, physics, computer science — all of them seemed super-cool. I desired to do all of it,” she says.
She settled on pursuing an undergraduate math degree but took all of the physics and computer science courses she could cram into her schedule.
Digging into data evaluation
After receiving a Marshall Scholarship, Broderick spent two years at Cambridge University in the UK, earning a master of advanced study in mathematics and a master of philosophy in physics.
Within the UK, she took numerous statistics and data evaluation classes, including her first-class on Bayesian data evaluation in the sphere of machine learning.
It was a transformative experience, she recalls.
“During my time within the U.K., I noticed that I actually like solving real-world problems that matter to people, and Bayesian inference was getting used in a few of crucial problems on the market,” she says.
Back within the U.S., Broderick headed to the University of California at Berkeley, where she joined the lab of Professor Michael I. Jordan as a grad student. She earned a PhD in statistics with a give attention to Bayesian data evaluation.
She decided to pursue a profession in academia and was drawn to MIT by the collaborative nature of the EECS department and by how passionate and friendly her would-be colleagues were.
Her first impressions panned out, and Broderick says she has found a community at MIT that helps her be creative and explore hard, impactful problems with wide-ranging applications.
“I’ve been lucky to work with a extremely amazing set of scholars and postdocs in my lab — sensible and hard-working people whose hearts are in the proper place,” she says.
Certainly one of her team’s recent projects involves a collaboration with an economist who studies the usage of microcredit, or the lending of small amounts of cash at very low rates of interest, in impoverished areas.
The goal of microcredit programs is to lift people out of poverty. Economists run randomized control trials of villages in a region that receive or don’t receive microcredit. They wish to generalize the study results, predicting the expected end result if one applies microcredit to other villages outside of their study.
But Broderick and her collaborators have found that results of some microcredit studies may be very brittle. Removing one or just a few data points from the dataset can completely change the outcomes. One issue is that researchers often use empirical averages, where just a few very high or low data points can skew the outcomes.
Using machine learning, she and her collaborators developed a way that may determine what number of data points should be dropped to alter the substantive conclusion of the study. With their tool, a scientist can see how brittle the outcomes are.
“Sometimes dropping a really small fraction of information can change the key results of an information evaluation, after which we would worry how far those conclusions generalize to recent scenarios. Are there ways we will flag that for people? That’s what we’re getting at with this work,” she explains.
At the identical time, she is constant to collaborate with researchers in a variety of fields, equivalent to genetics, to know the professionals and cons of various machine-learning techniques and other data evaluation tools.
Glad trails
Exploration is what drives Broderick as a researcher, and it also fuels one in every of her passions outside the lab. She and her husband enjoy collecting patches they earn by mountain climbing all the paths in a park or trail system.
“I feel my hobby really combines my interests of being outdoors and spreadsheets,” she says. “With these mountain climbing patches, you have got to explore every thing and then you definitely see areas you wouldn’t normally see. It’s adventurous, in that way.”
They’ve discovered some amazing hikes they’d never have known about, but in addition launched into greater than just a few “total disaster hikes,” she says. But each hike, whether a hidden gem or an overgrown mess, offers its own rewards.
And identical to in her research, curiosity, open-mindedness, and a passion for problem-solving have never led her astray.