177 Huntington Avenue
Boston, MA 02115
Benjamin A. Miller is a first-year PhD student at Northeastern’s Network Science Institute. His research is focused on the development and application of machine learning techniques that are robust to adversarial activity, particularly in the context of cybersecurity.
Prior to joining Northeastern, Ben was a technical staff member of the Cyber Analytics and Decision Systems Group at MIT Lincoln Laboratory. In this role, he developed novel algorithms for real-time linearization of radio-frequency electronics, researched methods and models for signal recovery in multisensor compressed sensing, and developed efficient spectral techniques for the detection of anomalies in large graphs. He continues his affiliation with the Laboratory as a Lincoln Scholar.
Ben is a senior member of the Institute of Electrical and Electronics Engineers (IEEE), and is an active member of the Association for Computing Machinery, the Society for Industrial and Applied Mathematics, and the IEEE Signal Processing Society. He holds seven patents and is author or coauthor of 39 peer-reviewed conference and journal papers on nonlinear signal processing, compressive sensing, and detection and estimation theory for graph-based data.
Ben received the BS degree (with highest honors) and the MS degree in computer science in 2005 from the University of Illinois at Urbana-Champaign.
- BS in Computer Science, University of Illinois at Urbana-Champaign
- MS in Computer Science, University of Illinois at Urbana-Champaign
- Hometown: Urbana, IL
- Field of Study: Network Science
- PhD Advisor: Tina Eliassi-Rad
What are the specifics of your graduate education (thus far)?
I got my MS degree 13 years ago and finally decided it’s time to go back to school! I’ve just started and am taking classes on machine learning and complex networks. I’m also getting involved in a new program on early detection of cyber attacks.
What are your research interests in a bit more detail? Is your current academic/research path what you always had in mind for yourself, or has it evolved somewhat? If so, how/why?
I’m very interested in contexts in which we can obtain provable performance guarantees for machine learning algorithms. This is important for determining how an adversary could impact performance. I’ve been interested for much of my career in what makes things “detectable,” and the recent academic work in adversarial machine learning and explainable machine learning are a big part of what encouraged me to pursue a PhD in the area.
What’s one problem you’d like to solve with your research/work?
If I could develop, for some common problem and common data source (e.g., detecting attacks via netflow logs), a “signal-to-noise”-like metric to quantify the difficulty of the problem, I would be extremely excited. In particular, I would be interested in any differences between theoretical and practical performance.
What aspect of what you do is most interesting/fascinating to you? What aspects of your research (findings, angles, problems you’re solving) might surprise others?
From my perspective, we’ve come to treat “analytics” as a black-box engine where we throw maximal computing resources at whatever question we have. I’m interested in resource constrained environments, where a user may have to make a choice between getting a mediocre answer quickly or a good answer later. Getting a better understanding the implications of this tradeoff is a big part of my interest.
What are your research/career goals, going forward?
I intend to continue my career as an applied researcher, focusing primarily on national security issues. I’d also like to mentor the next generation of computer scientists, perhaps by co-advising students.