Computational Neural Information Eng. Lab.
Austin J. Brockmeier
My research interests include data science, machine learning, signal processing, and the underlying mathematics and design of statistical models, optimizations, and algorithms. I have experience interfacing with complex data from the domains of biomedical engineering, neuroscience, and natural language processing.
I am an assistant professor in the Department of Electrical and Computer Engineering (ECE) and the Department of Computer and Information Sciences (CIS) within the College of Engineering at the University of Delaware (UD), and a resident faculty in UD's Data Science Institute.
306 Evans Hall
139 The Green
Newark, DE 19716
Select research outputs:
Quantifying the informativeness of similarity measurements, Journal of Machine Learning Research, 2017.
Learning shift-invariant waveforms to describe EEGs, IEEE Transactions on Biomedical Engineering, 2016.
Kernel-based metric learning for neural decoding, Neural Computation, 2014.
- Carlos Mendoza Cárdenas, M.S. spotting brain wave patterns using neural networks
- Yuksel Karahan, M.S. data science techniques for semisupervised domain transfer (Summer
- C. Cesar Claros Olivares, B.Eng., M.S. vigilant machine learning (Summer 2019–)
- Bilal Riaz, M.S. unsupervised learning as optimizations with positive semidefinite matrices (Fall 2019–)
- Hassan Baker, M.S. neural decoding and signal processing
- Hau Phan reinforcement learning estimation and detection
- Edwin Salcedo, M.Sc., M.B.A. machine learning for semisupervised domain transfer (Summer
The lab is open to both graduate and motivated undergraduate students.
The project focus should include machine learning (theory, algorithms,
and computation), data science (statistical analysis, information
retrieval, text mining, and visualization), and/or signal processing
(biomedical signals, audio, and remote sensing).
In the spring semester, I will
be teaching Signals and Systems (ELEG 305). This course is a meaningful
pre/co-requiste for undergraduate research with me. Another course to
take is ELEG 310 (Random Signals and Noise).
Graduate students should have experience in mathematical modeling or analysis, numerical programming, and
statistics and have an interest in the theoretical, computational, or
algorithmic research aspects. I teach Large Scale Machine Learning
(ELEG/FSAN 817) in the fall semester.
Students could also aim for the data science triad of statistics, computation, and domain
expertise. These students should have a background in an analytical science, engineering, or medicine and
have an interest in using machine learning, signal processing, and data science to pursue a scientific or
engineering research project. There are opportunities to co-advise students interested robotics or
neuroscience. One of the primary goals of the lab is developing new techniques for processing and analyzying
data from the central or peripheral nervous system using
electrophysiological recordings (EEG, EMG, ECoG) and other brain imaging
technology (fMRI, NIRS).
Information on admission and funding.
I was a research fellow at the National Centre for Text Mining within the School of Computer Science at the University of Manchester, UK. My mentor was Prof. Sophia Ananiadou. Before that I was at the University of Liverpool, where my mentor was Prof. John (Yannis) Goulermas. At both universities, I contributed to a project funded by the UK's Medical Research Council that developed techniques that can assist the literature screening stage of systematic reviews of literature, with public health as the main focus.
I received a Ph.D. in Electrical and Computer Engineering from the University of Florida. My advisor was Prof. Jose C. Principe. As a Ph.D. student I developed algorithms for extracting relevant patterns from signals and applied them to the analysis of multiscale neural data, both action potential timings (spike trains) and electrical potentials (local field potential and EEG recordings). We collaborated with researchers on sensory processing and brain machine interfacing.
I received a B.S. degree in computer engineering from the University of Nebraska–Lincoln (UNL) at its Omaha campus, having received the Walter Scott, Jr. Scholarship and the University of Nebraska Regents Scholarship. In Omaha, UNL's engineering departments are housed within the Peter Kiewit Institute, while my second major in mathematics and minor in computer science are from the respective departments at the University of Nebraska at Omaha.
I graduated from Cozad High School, and was honored to have received the Rick Wilson Undergraduate Scholarship. Among many activities, one of the most formative was my participation in the Nebraska Academic Decathlon.