Gabriele Farina grew up in a compact town in the hilly wine region of northern Italy. Neither of his parents had a college degree, and although they both believed they “didn’t understand math,” Farina says, they bought him the technical books he asked for and didn’t discourage him from attending a science-oriented high school rather than a classical high school.
Around the age of 14, Farina focused on an idea that would prove to be the foundation of his career.
“I was fascinated very early on by the idea that a machine could predict and make decisions much better than humans,” he says. “The fact that human-made mathematics and algorithms can create systems that are in some ways superior to their creators, while still relying on simple elements, has always been a major source of awe for me.”
At age 16, Farina wrote code for a board game he played with his 13-year-old sister.
“I used game by game to calculate the optimal move and prove to my sister that she had already lost long before either of us could see it with our own eyes,” Farina says, adding that his sister was less thrilled with his modern system.
Currently, as an assistant professor in MIT’s Department of Electrical Engineering and Computer Science (EECS) and principal investigator in the Laboratory for Information and Decision Systems (LIDS), Farina combines concepts from game theory with tools such as machine learning, optimization, and statistics to advance the theoretical and algorithmic foundations of decision making.
Starting her studies at the Politecnico di Milano, Farina studied automation and control engineering. Over time, however, he realized that what sparked his interest was not “just applying known techniques, but understanding and expanding their fundamentals,” he says. “Gradually I turned more and more towards theory, but I was still very concerned with demonstrating concrete applications of that theory.”
Farina’s advisor at the Politecnico di Milano, Nicola Gatti, a professor and researcher in computer science and engineering, introduced Farina to research topics in computational game theory and encouraged him to pursue a Ph.D. The first in his immediate family to earn a college degree and living in Italy, where doctorates are treated differently, Farina says he didn’t even know what a doctorate was.
Nevertheless, a month after completing his bachelor’s degree, Farina began a PhD program in computer science at Carnegie Mellon University. There, he won honors for his research and dissertation, as well as a Facebook Fellowship in Economics and Computation.
While completing her PhD, Farina worked for a year as a research associate in Meta’s basic artificial intelligence research labs. One of his major projects was helping develop Cicero, an artificial intelligence that could beat human players in a game of forming alliances, negotiating, and detecting when other players were bluffing.
Farina says, “When we built Cicero, we designed him so that he wouldn’t agree to an alliance if it wasn’t in his interest, and similarly understood if a player was likely to lie because following through on his proposal would go against his own incentives.”
A 2022 paper in the aforementioned Cicero could mark progress toward artificial intelligence that can solve convoluted problems that require trade-offs.
After a year at Meta, Farina joined the MIT faculty. In 2025, he was awarded the CAREER Award of the National Science Fund. His work – based on game theory and its mathematical language of describing what happens when different parties have different goals and then quantifying an “equilibrium” where no one has a reason to change their strategy – aims to simplify huge, convoluted real-world scenarios in which such an equilibrium could take a billion years to calculate.
“I’m researching how we can use optimization and algorithms to actually find these stable points effectively,” he says. “Our work attempts to shed new light on the mathematical foundations of the theory, provide better control and prediction of these complex dynamic systems, and uses these ideas to compute good solutions to large interactions between multiple agents.”
Farina is particularly interested in environments with “imperfect information”, meaning that some agents have information unknown to other participants. In such scenarios, information has value, and participants must strategically approach the information they have so as not to reveal it and reduce its value. An everyday example occurs in the game of poker, where players bluff to hide information about their cards.
According to Farina, “we currently live in a world where machines are much better at bluffing than humans.”
The “enormous amount of imperfect information” situation took Farina back to the origins of the board game. Stratego is a military strategy game that has inspired a research effort that has cost millions of dollars to create systems that can defeat human players. Requiring a convoluted calculation of risk and misrepresentation or bluffing, it was perhaps the only classic game where the greatest efforts did not produce superhuman results, Farina says.
Thanks to modern algorithms and training costing less than $10,000 rather than millions, Farina and his research team were able to beat the greatest player of all time – by 15 wins, four draws and one loss. Farina says he’s excited to be able to obtain such results in such a cost-effective way and hopes “these new techniques will be incorporated into future pipelines,” he says.
“We’ve seen continued progress in building algorithms that can reason strategically and make sound decisions despite large operating spaces or imperfect information. I’m excited about the opportunity to incorporate these algorithms into the broader AI revolution happening around us.”