When You Think of AI, Think of Stevens: Relaunching the Stevens Institute for Artificial Intelligence
Brendan Englot’s vision for the relaunch of the SIAI research center prioritizes increased research, education and advocacy to solidify Stevens’ reputation for AI and machine learning expertise worldwide
Self-cleaning litter boxes. Driverless package delivery. Digital assistants. ChatGPT. Drugs invented by algorithms. Job interviews conducted by chatbots. Watches that recognize a heart attack — and call 9-1-1.
Artificial intelligence (AI) and machine learning have quickly evolved to alter everyday life in ways that would be unrecognizable outside of popular fiction just 15 years ago.
With the very nature of human life being changed forever by these technologies, the Stevens Institute for Artificial Intelligence (SIAI) finds itself in a stage of rapid evolution.
Launched in 2018, Stevens Institute of Technology’s SIAI is an interdisciplinary collaboration of researchers representing more than 100 faculty from all four of the university’s academic units: the Schaefer School of Engineering and Science; the School of Business; the School of Systems and Enterprises; and the School of Humanities, Arts and Social Sciences.
Together, these faculty are dedicated to conducting research and outreach efforts that address pressing global problems through AI and machine learning technologies.
“At its core, SIAI is a grouping of faculty from very diverse disciplines who have a shared interest in AI, coming together and achieving an impact where our collective work can be more effective and wide-reaching than our own individual research activities,” said SIAI Director Brendan Englot. Englot is also the Geoffrey S. Inman Junior Professor and an associate professor in the Department of Mechanical Engineering.
Representing one of the university’s six foundational research pillars, SIAI partners with industry, government, foundations and academic institutions to develop solutions to urgent issues in such areas as healthcare, education, business and national security, while simultaneously pioneering new AI and machine learning technology and scholarship.
The research center is also unusually big.
“SIAI is unique because of how broad its membership is,” said Englot. “About one-third of the entire body of full-time faculty at Stevens are members.”
To better serve this growing membership and the university as a whole, SIAI will relaunch on Wednesday, October 18, 2023, with a new location, new goals and a new plan for success. The relaunch will also coincide with the first event of the 2023-24 Schaefer School Dean’s Lecture series, featuring JPMorgan Chase Head of AI Research Manuela Veloso.
Englot’s goal for SIAI is to further establish the university as a global resource for AI and machine learning expertise, ultimately making Stevens’ name as synonymous with AI and machine learning as it has been with engineering for the last 150 years.
To do so, Englot’s vision for the new SIAI prioritizes facilitating larger, more ambitious research efforts; increased education and outreach activities; and raising and solidifying Stevens’ profile and voice in the global conversation about the ethical uses and applications of machine learning and AI.
SIAI as incubator and force multiplier
Englot’s topmost priority for SIAI's next phase is “to serve as an incubator” to help its members pursue more and more expansive collaborative research initiatives in AI and machine learning.
“We would really like to see a measurable change in research activity resulting from this relaunch,” said Englot. “I would love to see more successful large-scale multidisciplinary collaborations that could use SIAI as a force multiplier, where three or four faculty come together and submit a proposal for multimillion-dollar research.”
The necessary pieces for undertaking such efforts are already in place, Englot said, including top talent on faculty representing a wide array of diverse disciplines.
“We already have AI research as a strength at Stevens at the grassroots level: It's just distributed amongst a lot of faculty working in their own labs on their own applications,” he said. “I think bringing us together through SIAI will unlock our full potential.”
Even Stevens’ physical location — centrally located in the New York City metropolitan area — makes it perfectly situated to take advantage of the region’s many unique AI-related opportunities for collaboration, including the tech and healthcare sectors in New York and the biotech and pharmaceutical sectors in New Jersey.
Englot’s goal is to develop new and additional resources to assist faculty through the long, often complicated proposal process.
Within the past year, proposals for more ambitious AI and machine learning initiatives have already increased with SIAI’s help. Englot himself, in collaboration with mechanical engineering Assistant Professor Hamid Jafarnejad Sani, was recently awarded a large contract to develop an algorithm-based software system that creates photorealistic 3D site models from images taken at varying altitudes.
SIAI as a (physical) place of inspiration
Since its launch five years ago, SIAI has existed as a research center in the intellectual sense. But with its relaunch in October, SIAI will unveil its first permanent physical space.
Located on the fourth floor of the Gateway Academic Center north building, the new SIAI headquarters will include faculty offices, a conference room and a new high-performance computing cluster for faculty and Ph.D. students. SIAI will also now employ a full-time operations manager dedicated to supporting SIAI faculty.
Englot wants to develop SIAI in general and this new space specifically into a place of opportunity, where people from different disciplines, departments and schools can interact in both formal and informal ways. By increasing opportunities to convene, Englot hopes SIAI can help spark new collaborative partnerships, ideas and innovations.
“We hope our physical space will facilitate more conversations and allow faculty to get to know all of their different neighbors — to learn the many things they have in common that they may not know they have in common,” he said.
In Spring 2023, SIAI hosted a networking lunch to facilitate such conversations, in which members were invited to briefly introduce themselves and their research interests. Faculty from all four schools attended, allowing them to learn from each other across disciplines in a casual and sociable environment.
“The reaction we got to that was very positive. People were enthusiastic to have an open invitation to just come and exchange ideas,” Englot said.
Englot plans to increase the number of such events and opportunities available both inside and outside the new SIAI facilities. Additionally, he hopes the new space will allow SIAI to expand its educational outreach activities and increase opportunities to host student and community tours of the university’s AI research labs.
SIAI for student development
Englot’s vision for SIAI is not limited to engagement opportunities for faculty. Rather, he would like to increase opportunities for university students at all levels to share their voice in the AI conversation and to help them unearth opportunities they might not have known about otherwise.
“With all those faculty coming together, we will also bring students into the process. We want to have more ways that they can get involved in multidisciplinary research with us,” Englot said.
One part of this effort is continuing to sponsor the AI Research in Summer (AIRS) Fellowship Program, in which students are partnered with faculty mentors from all four Stevens schools to gain firsthand experience in AI- and machine learning-related research projects.
While many fellowship programs are restricted to advanced graduate students, the AIRS fellowship is unique, said Englot, because it is open to undergraduate and master’s students specifically. A poster session highlighting research conducted by the 2023 AIRS student cohort will be held in the fall.
Although the university hosts a number of AI-related programs and projects, many of them sit within individual schools, departments or concentrations. Englot would like to build SIAI into a one-stop shop where students can easily connect to opportunities and resources beyond their immediate course of study.
“We want SIAI to serve as a core where all of the university’s intellectual activity in AI can be housed,” he said. “We can point [students] in the right direction, inform them about relevant seminars and programs, and connect them with the right faculty members.”
Englot also plans to increase the number of research seminars featuring leading thought leaders in AI.
SIAI evolving
In addition to celebrating a new phase of SIAI’s development during the fall’s upcoming relaunch events, SIAI, under Englot’s leadership, will continue to examine its approach so it remains relevant in the ever-changing AI landscape.
One question currently on Englot’s mind is whether SIAI’s organizational structure, first implemented when the center was established, remains appropriate for SIAI’s future.
SIAI’s members are currently organized into nine research clusters, each representing a different focus area of AI-related research.
“One of the things that I think we may benefit from is considering whether the nine clusters are still the best description of what we do and what our capabilities are,” Englot said.
An alternative under discussion involves eliminating the research clusters altogether and replacing them with an organizational structure in which SIAI members are instead categorized as either core or affiliate members.
“Those core members would have a real stake in the ownership and the success of SIAI and maybe additional access to resources as well,” Englot said.
This reorganization of operations management would directly affect SIAI’s future opportunities and ability to support the university as a whole.
“As more faculty get involved with SIAI, more of the revenue from their research expenditures will come back to SIAI, which will allow us to grow, hire more staff, expand our facilities and make more computing resources available,” Englot explained. “With more involved members who are really dedicated stakeholders, their success will mean we can better support everyone's research activities in AI.”
SIAI as advocate, educator and agent of change
The ways in which AI has the potential to benefit society are innumerable, from simple conveniences like Google Maps to advanced diagnostic and treatment tools in healthcare and forecasting tools for combating climate change.
But SIAI’s strengths lie not only in what its researchers can accomplish directly, but also in its ability to shine a light on the misuses of AI that result in harm, such as disinformation and social manipulation, algorithmic discrimination, copyright and privacy infringement, and cybercrime.
“One of the things that academics are in the best position to do is to ask tough questions — to hold big players in this space accountable,” Englot said. “We can be an important voice in that conversation, push for improved transparency and try to help steer AI toward ethical use.”
Having a voice in this global conversation also means positioning SIAI as a trusted source for AI-related information for the general public, said Englot. The growing prevalence of AI technologies, in fact, is a topic the public has shown to have increasing anxiety about.
A recent national TechPulse survey on public sentiment on AI conducted on behalf of Stevens revealed that U.S. adults are 10% less likely in 2023 than in 2021 to believe that the positives of AI in everyday life outweigh the negatives.
Of the 2200 adults surveyed nationwide, 44% described negative emotions like concern, anxiety or fear as the strongest emotion they feel when thinking about AI.
Although personally optimistic about AI, Englot understands Americans’ wariness. As powerful as AI tools are, he noted, the lack of transparency in generative AI tools like ChatGPT, for example, justify cause for concern.
By increasing SIAI's reach as an established authority on both the potential benefits and potential pitfalls of AI and AI-related technologies, he hopes to help assuage the public’s fears while making them more informed consumers, citizens and decision-makers.
SIAI as identity
In addition to serving as SIAI director, Englot is an established expert on mobile robots and autonomous navigation for unmanned underwater, ground and aerial vehicles. He has received the Office of Naval Research Young Investigator Award, the National Science Foundation CAREER Award and the Stevens Provost's Award for Research Excellence, and has been awarded $4.4 million in research funding as a principal investigator. Named by AMiner as one of the top 100 most cited roboticists in the world, Englot joined the Stevens faculty in 2014.
In his fifteen-year career in mechanical engineering, Englot has personally witnessed how quickly and dramatically AI has altered his chosen research area and the world at large.
“I've worked on robotics from the standpoint of how an autonomous mobile robot or an autonomous vehicle sees the world, understands the world, makes decisions and navigates,” he explained. “Originally, that discipline did not involve any artificial intelligence, but, like with so many others, AI has been a disrupter and a game-changer, allowing us to build much more intelligent and capable autonomous vehicles. That's how I got into the AI space.”
Englot joined SIAI when it launched as the coordinator for the robotics, perception and human-machine interaction research cluster. He stepped up to serve as SIAI’s interim director in 2022 before taking the position full-time in January 2023.
Although SIAI’s efforts were impeded by the COVID-19 pandemic starting in 2020, Englot said he’s glad to be helping to rebuild the strong momentum that the center had when it first launched. Ultimately, he would like to see Stevens become as famous for AI and machine learning as it is in other major research areas.
“AI as a strength isn't as well-known as some of the other long-established centers we've been more historically known for, like the work in the that's been going on for decades,” said Englot. “With this relaunch, SIAI can be the face of AI for Stevens and improve our external visibility, so that when you think of Stevens, one of the top three things you think of is our capabilities in AI.”