Dr. Eric Braude
Dr. Eric Braude Shares Professional Plaudits with Student Research Assistants, Is Proud of Their Accomplished Careers
Associate Professor and Director of Digital Learning, Computer Science
PhD, Columbia University; MS, University of Miami; MS, University of Illinois; BSc, University of Natal
What is your area of expertise?
Machine learning and software engineering.
Tell us about your work—can you share any current research or recent publications?
A paper, “Generalizing Morley’s and Other Theorems with Automated Realization,” by my MS in Computer Science alum Satbek Abdyldayev (MET’17) and me, appeared in the Journal of Automated Reasoning. In this 23-page paper, we describe the theory behind as well as our GEOPAR program, which checks the validity of various proposed theorems in plane geometry. Researchers at Dartmouth and Princeton recently commented on this work, calling us “champions” of the approach.
In addition, MS in Software Development graduate Jason Van Schooneveld (MET’18) and I published “Incremental UML for Agile Development with PREXEL,” which was introduced at the International Conference on Software Engineering in Gothenburg, Sweden.
How does your work apply in practice? What is its application?
RUML helps to solve the problem of making comprehensible large software designs, consisting of hundreds of parts. We also have ongoing projects that apply technology to learning and evaluation, areas of longstanding interest to all of us. Student, Laura Kocubinski, is beginning her MS thesis on the analysis of news articles using machine learning.
What courses do you teach at MET?
Most recently, Machine Learning (MET CS 767), Artificial Intelligence (MET CS 664), Advanced Programming Techniques (MET CS 622), Analysis of Algorithms (MET CS 566), and Systems Analysis and Design (MET CS 682).
You taught Machine Learning (MET CS 767) online for Fall 2. For those interested in studying online, can you please talk about the online classroom experience?
Online learning can be very effective for adults. A surprising amount can be learned in seven weeks, and we ensure that our online courses are just as demanding as our 14-week on-campus versions. The value has a lot to do with concentration, students’ ability to replay recorded classes (including at high speed), and to get help and ask questions via voice, video, or text in real time and asynchronously.
Please highlight a particular project within your courses that most interests your students.
In my Machine Learning, Artificial Intelligence (AI), and advanced programming courses, students select their own projects in addition to weekly labs. They align their selection with their interests, their job, or a data source they have identified. There are a huge variety of data sources available now, on every conceivable topic. One sample project is an application that learns to play expert Tetris by playing many games and evaluating itself. Another project matched grant sources with applicants. Many aspects of society are represented.
Are there opportunities for fruitful collaboration with faculty, or even students, in terms of marrying practice with research activities? If so, can you provide an example?
Yes, numerous students have been involved in our projects. Besides Aviral Srivastava, at least three other students have been involved in the RUML project. At least four have participated in the JGrams project.
What is the role of an RA? For example, how did you collaborate with Aviral in this role?
Typically, I initiate an idea, discuss it thoroughly with the RA, then modify or replace it if needed. We do a literature search, and the RA writes a program. Aviral has implemented a key part of RUML that my colleague Michael Huang and I have been developing for about two years.
From your previous work in the industry, what “real-life” exercises do you bring to class? And how does that inform your classroom?
Before coming to BU, I worked on applying AI to complex systems, and on software reliability. These projects, together with consulting, continue to influence my research and teaching. A memorable consulting task was for the Canadian Navy, on how too much redundancy can degrade a system. I have done basic research and found it stimulating, but doing research that affects the real world provides a check that the work is going in a productive direction.
Is there anything else you would like to add?
I am proud of the students who have contributed to research with me. They have gone on to industry positions, entrepreneurial ventures, doctorates, and university teaching.