UD Seal
Prof. John Cavazos


Prof. John Cavazos

Welcome to my personal web site.

I am an associate professor in the Computer and Information Sciences Department and previously JP Morgan Chase Faculty Fellow in the Institute for Financial Services Analytics. I also have a joint appointment in the Electrical and Computer Engineering Department. My research interests are in high-performance computing, machine learning, predictive analytics, and application of these technologies to hard problems.

I was one of the first researchers to work on applying machine learning to improve compilers. This field of research has rapidly grown to include hundreds of authors from many major universities and technology companies. Compilers typically contain many heuristics to solve hard problems approximately and efficiently. Finding heuristics that perform well on a broad range of applications and processors is one of the most complex tasks faced by compiler writers. My research involves using machine learning techniques to automatically construct compiler optimization heuristics. I have shown that this technique can completely eliminate the human from heuristic design. My research on applying machine learning to compiler optimizations received the NSF CAREER award. My current research focus is on large-scale static and dynamic analysis of malicious applications.

Selected Cybersecurity Publications:

  • Large-scale exploration of feature sets and deep learning models to classify malicious applications. Tristan Vanderbruggen and John Cavazos. IEEE Resilience Week (RWS) 2017
  • Improving the effectiveness and efficiency of dynamic malware analysis with machine learning. Sean Kilgallon, Leonardo De La Rosa, and John Cavazos.IEEE Resilience Week (RWS) 2017
  • Parallelization of Machine Learning Applied to Call Graphs of Binaries for Malware Detection. Robert Searles, Lifan Xu, William Killian, Tristan Vanderbruggen, Teague Forren, John Howe, Zachary Pearson, Corey Shannon, Joshua Simmons, John Cavazos. PDP 2017
  • HADM: Hybrid Analysis for Detection of Malware. Lifan Xu, Dongping Zhang, Nuwan Jayasena, John Cavazos. IntelliSys 2016 [PDF]


  • Software Automatic Tuning: From Concepts to State-of-the-Art Results.
  • Editors : Ken Naono, Keita Teranishi, John Cavazos, and Reiji Suda. Springer 2011

Benchmark Suites

  • FinanceBench This benchmark suite is aimed at those who work with financial code to see how certain code paths can be targeted to accelerators.
  • PolyBench/ACC Is a collection of benchmark kernels that exhibit parallelism and have been ported to accelerators using a variety of different parallel languages.

This research was generously sponsored by the National Science Foundation, DARPA, JP Morgan Chase, Army, Naval Research Laboratory, and Google.

My Students