SEAS

Reimagining STEM Learning Objectives in Response to Generative AI


image of Vijay Janapa ReddiVijay Janapa Reddi, Associate Professor (SEAS) and director of the Edge Computing Lab, is an applied machine learning computer architect. As a scholar with deep knowledge of how artificial intelligence (AI) works, Janapa Reddi offers a unique perspective on both the challenges and opportunities generative AI presents. Generative AI platforms, such as ChatGPT, are changing how students interact with course material and setting new standards for the skills necessary for future professional fields. While Janapa Reddi is cautious about implementing exercises that leverage such platforms in his COMPSCI 141: Computing Hardware course, he suggests that faculty seize this moment to reevaluate their teaching objectives and consider how they can support students to develop the skills they will need to navigate and use these new technologies in their careers. Imagine a future where every engineer is supported by a personalized AI assistant, offering guidance throughout their processes, enabling them to design optimal, robust, secure, and highly efficient systems.

Leveraging digital spaces to enhance student engagement


image of Rebecca and Charles NessonRebecca Nesson, Dean for Academic Programs, SEAS, and Charles R. Nesson, William F. Weld Professor of Law at Harvard Law School, Founder of the Berkman Klein Center for Internet & Society, and Principal Investigator of BKC’s Nymity project, have worked together for many years as a unique father-daughter teaching team at Harvard. Ranging from First-Year Seminars to offerings at the Law School and the Extension School, their courses focus on the deliberative practices of juries and their role in determining justice. Since 2006, the Nessons have embraced new technology in their classrooms to encourage student engagement and productive dialogue across differences as they and their students consider issues of jury bias and power in their courses. 

Leveraging asymmetry in student's prior knowledge through peer learning exercises


Salil VadhanSalil Vadhan, Vicky Joseph Professor of Computer Science and Applied Mathematics & Lead PI on Harvard’s Privacy Tools Project, teaches COMPSCI 120: Introduction to Algorithms and their Limitations, a new introductory course in theoretical computer science “aimed at giving students the power of using mathematical abstraction and rigorous proof to understand computation with confidence.” Many computer science students are “builders” who enjoy the creative aspect of the field, yet their mathematical backgrounds are often quite diverse; to some, mathematical theory is unfamiliar. In redesigning the undergraduate computer science curriculum, it was a priority to make this “new language, reasoning, and way of thinking” accessible to students early in the program.... Read more about Leveraging asymmetry in student's prior knowledge through peer learning exercises

Designing solutions in response to real-world problems


Martin BechtholdMary TolikasMartin Bechthold, the Kumagai Professor of Architectural Technology at the Harvard School of Design, and Mary Tolikas, Chief Innovation Officer at the Dana Farber Cancer Institute and Visiting Lecturer on Engineering Sciences at the Harvard John A. Paulson School of Engineering & Applied Sciences, work together to teach Independent Engineering Design Project I & II, a small, seminar-size course for second-year students in the Master in Design Engineering program. Students spend the academic year designing prototype solutions to real-world problems about which they are passionate. Throughout the two semesters, students work towards understanding the complexity and dynamics underlying the problem and, by collaborating with relevant stakeholders, they explore visionary solutions and iterate on prototypes that would best address their challenge. Analyzing and quantifying potential impact is central to solution development. Early in the course, students are placed in self-selected “affinity groups,” based on shared interests. Students use these groups throughout the term to bounce ideas from and relay feedback. They also receive regular feedback from faculty advisors throughout each stage as they continue to evolve their project.

Teaching system-level thinking with an interdisciplinary lens


Fawwaz HabbalDoris SommerFawwaz Habbal, Executive Dean for Education and Research and Senior Lecturer on Applied Physics, and Doris Sommer, Ira Jewell Williams, Jr. Professor of Romance Languages and Literatures and of African and African American Studies, co-teach systems-level thinking. Their course, Aesthetic Pleasure and Smart Design: Janus Faces the Future, trains students to look at complex problems from the perspective of both artists and engineers. This requires the development of skills in scientific assessment and disinterested aesthetic judgment. In the spirit of Renaissance Now, an international movement to promote sustainable development, Habbal and Sommer model the combination of boldness and humility. Students in ES 27 read and reflect on material which ranges from aesthetic philosophy and history to triggers for scientific revolution. Then they tackle a complex problem through a proposal that will gain aesthetic acceptance and be scientifically effective. 

Encouraging learning by creating alongside diverse feedback


Paul BottinoPaul B. Bottino, Co-Founder, Executive Director, and Lecturer at the Technology & Entrepreneurship Center at Harvard, offers Start-up R&D to undergraduate students across disciplines who are interested in the field and have a particular project idea in mind. Within the workshop course structure, “each student project is the educational centerpiece.” Student groups work on a variety of innovative startup projects seeking solutions to problems they care about. The course uses multiple approaches to help students build upon their ideas and receive constructive feedback: “challenge sessions” where students outline their biggest obstacles to a small group of peers; individual meetings with Bottino and teaching fellows; and connections with alumni. “It’s like a Greek forum of peers, near-peers, and mentors” with students learning that “entrepreneurship is a creative and iterative research practice of idea formulation, experimentation, and feedback.” At the end of term, students present and receive feedback on projects at a public event “Demo Day.”

Flipping the classroom for deeper student engagement and feedback on learning


L MahadevanL Mahadevan, Lola England de Valpine Professor of Applied Mathematics in SEAS, and Professor of Organismic and Evolutionary Biology, and of Physics in FAS used a 2017-2018 SEAS Learning Incubator LInc Faculty Fellowship to emphasize active learning in his Mathematical Modeling course. He implemented a flipped classroom approach to enable students to come to class with problems and questions to collaborate on, time to develop their own problems from scratch, and work on modeling with peers. The foundational arc supporting this process has students move from observations through abstraction, analysis and communication, and iteration.

Simple examples lead to deep engagement


Scot MartinThree years ago, Scot T. Martin, Gordon McKay Professor of Environmental Science and Engineering and Professor of Earth and Planetary Sciences, decided to “start from scratch” with his approach to teaching thermodynamics. In his course Thermodynamics by Case Study, he found that by focusing on every day, concrete examples (e.g., running, the function of the heart) and demanding an intense level of participation, he could help students unpack layer after layer to rediscover and truly understand the fundamental laws. 

Real problems: Teaching theory through practice


Jelani Nelson: Profile PhotoJelani Nelson, Assistant Professor of Computer Science, assigns students real programming problems in his introductory algorithm courses, CS124 Data Structures and Algorithms and CS125 Algorithms & Complexity. Students write and test their coded solutions to practice problems via an open server on the course website and receive immediate feedback on their work.

The hiccups, humility, and benefits of deciding to flip a course


Margo SeltzerMargo Seltzer, Herchel Smith Professor of Computer Science, flipped part of her course, CS161, "Operating Systems."

What is a flipped classroom? Reversing the typical lecture and homework components of a course – students engage with video lectures/online materials in advance to free up class time for more experiential or hands-on activities.