Recent Ph.D. graduate, Thomas Campbell, looks forward to “doing impactful work for years to come,” in his new position at a Biotechnology company. Campbell is using his deep interest in machine learning and analysis to develop algorithms and machine learning methods for use in cancer diagnostic tests, as well as performing analyses in collaboration with academic and industrial researchers using the methods the group develops.
After exploring different career paths Campbell grew more interested in the machine learning methods. He transitioned into industry after spending a year a CU working as a postdoctoral researcher developing machine learning for event reconstruction in the proposed DUNE near detector.
Currently, Campbell works on a set of algorithms that use a very creative implementation of concepts like bagging and boosting to form an ensemble of weak learners to robustly classify incredibly high feature dimension, noisy, small data sets common in cancer (and other types of medical) research. The research group, Campbell states, “has made some real interesting progress in using machine learning for the low statistic, high feature dimension data sets common in many fields of medical research.”
When discussing his interest in his work Campbell says “The aspects of machine learning I find most interesting are generally finding novel and non-trivial applications of machine learning. Anyone can throw a gigantic image set at a neural net and get some decent results. I am more interested in specialized algorithms and more non-standard applications of machine learning.”
Campbell recounts enjoying his time at CSU. “With a relatively smaller class size, I felt quite connected with my fellow students. There was always collaboration to get assignments done and I never felt like we were pitted against one another.”
He also enjoyed working with Walter Toki on the T2K experiment, “The CSU High Energy Physics group was awesome too. Working on T2K, I enjoyed the opportunity to travel to Japan around a dozen times for meetings, data taking, and detector maintenance.” Campbell’s experience at CSU sparked a passion for data analysis and for trying to measure things we find difficult to measure. At CSU, Campbell received, “a solid foundation in data analysis and statistics for scientific research.”