The early years of college members’ careers are a formative and exciting time wherein to ascertain a firm footing that helps determine the trajectory of researchers’ studies. This includes constructing a research team, which demands revolutionary ideas and direction, creative collaborators, and reliable resources.
For a gaggle of MIT faculty working with and on artificial intelligence, early engagement with the MIT-IBM Watson AI Lab through projects has played a crucial role helping to advertise ambitious lines of inquiry and shaping prolific research groups.
Constructing momentum
“The MIT-IBM Watson AI Lab has been hugely necessary for my success, especially after I was starting out,” says Jacob Andreas — associate professor within the Department of Electrical Engineering and Computer Science (EECS), a member of the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), and a researcher with the MIT-IBM Watson AI Lab — who studies natural language processing (NLP). Shortly after joining MIT, Andreas jump-started his first major project through the MIT-IBM Watson AI Lab, working on language representation and structured data augmentation methods for low-resource languages. “It really was the thing that allow me launch my lab and begin recruiting students.”
Andreas notes that this occurred during a “pivotal moment” when the sector of NLP was undergoing significant shifts to grasp language models — a task that required significantly more compute, which was available through the MIT-IBM Watson AI Lab. “I feel just like the type of the work that we did under that [first] project, and in collaboration with all of our people on the IBM side, was pretty helpful in determining just tips on how to navigate that transition.” Further, the Andreas group was capable of pursue multi-year projects on pre-training, reinforcement learning, and calibration for trustworthy responses, because of the computing resources and expertise inside the MIT-IBM community.
For several other faculty members, timely participation with the MIT-IBM Watson AI Lab proved to be highly advantageous as well. “Having each mental support and in addition with the ability to leverage a few of the computational resources which can be inside MIT-IBM, that’s been completely transformative and incredibly necessary for my research program,” says Yoon Kim — associate professor in EECS, CSAIL, and a researcher with the MIT-IBM Watson AI Lab — who has also seen his research field alter trajectory. Before joining MIT, Kim met his future collaborators during an MIT-IBM postdoctoral position, where he pursued neuro-symbolic model development; now, Kim’s team develops methods to enhance large language model (LLM) capabilities and efficiency.
One factor he points to that led to his group’s success is a seamless research process with mental partners. This has allowed his MIT-IBM team to use for a project, experiment at scale, discover bottlenecks, validate techniques, and adapt as obligatory to develop cutting-edge methods for potential inclusion in real-world applications. “That is an impetus for brand new ideas, and that’s, I believe, what’s unique about this relationship,” says Kim.
Merging expertise
The character of the MIT-IBM Watson AI Lab is that it not only brings together researchers within the AI realm to speed up research, but in addition blends work across disciplines. Lab researcher and MIT associate professor in EECS and CSAIL Justin Solomon describes his research group as growing up with the lab, and the collaboration as being “crucial … from its starting until now.” Solomon’s research team focuses on theoretically oriented, geometric problems as they pertain to computer graphics, vision, and machine learning.
Solomon credits the MIT-IBM collaboration with expanding his skill set in addition to applications of his group’s work — a sentiment that’s also shared by lab researchers Chuchu Fan, an associate professor of aeronautics and astronautics and a member of the Laboratory for Information and Decision Systems, and Faez Ahmed, associate professor of mechanical engineering. “They [IBM] are capable of translate a few of these really messy problems from engineering into the form of mathematical assets that our team can work on, and shut the loop,” says Solomon. This, for Solomon, includes fusing distinct AI models that were trained on different datasets for separate tasks. “I believe these are all really exciting spaces,” he says.
“I believe these early-career projects [with the MIT-IBM Watson AI Lab] largely shaped my very own research agenda,” says Fan, whose research intersects robotics, control theory, and safety-critical systems. Like Kim, Solomon, and Andreas, Fan and Ahmed began projects through the collaboration the primary yr they were capable of at MIT. Constraints and optimization govern the issues that Fan and Ahmed address, and so require deep domain knowledge outside of AI.
Working with the MIT-IBM Watson AI Lab enabled Fan’s group to mix formal methods with natural language processing, which she says, allowed the team to go from developing autoregressive task and motion planning for robots to creating LLM-based agents for travel planning, decision-making, and verification. “That work was the primary exploration of using an LLM to translate any free-form natural language into some specification that robot can understand, can execute. That’s something that I’m very pleased with, and really difficult on the time,” says Fan. Further, through joint investigation, her team has been capable of improve LLM reasoning — work that “can be inconceivable without the IBM support,” she says.
Through the lab, Faez Ahmed’s collaboration facilitated the event of machine-learning methods to speed up discovery and design inside complex mechanical systems. Their Linkages work, as an illustration, employs “generative optimization” to unravel engineering problems in a way that’s each data-driven and has precision; more recently, they’re applying multi-modal data and LLMs to computer-aided design. Ahmed states that AI is steadily applied to problems which can be already solvable, but may gain advantage from increased speed or efficiency; nevertheless, challenges — like mechanical linkages that were deemed “almost unsolvable” — are actually within sight. “I do think that is certainly the hallmark [of our MIT-IBM team],” says Ahmed, praising the achievements of his MIT-IBM group, which is co-lead by Akash Srivastava and Dan Gutfreund of IBM.
What began as initial collaborations for every MIT faculty member has evolved into an enduring mental relationship, where each parties are “excited in regards to the science,” and “student-driven,” Ahmed adds. Taken together, the experiences of Jacob Andreas, Yoon Kim, Justin Solomon, Chuchu Fan, and Faez Ahmed speak to the impact that a durable, hands-on, academia-industry relationship can have on establishing research groups and bold scientific exploration.
