We are a group of experimental particle physicists interested in the application of modern machine learning (ML) techniques to analyze experimental physics data. Our research focus include geometrical pattern recognition for particle imaging detectors, understanding data/simulation discrepancies in algorithm response, and estimating uncertainties in ML applications.
We welcome any researcher (physicist or non-physicist) to join our group and share the joy and struggle of problem solving in general. Please feel free to contact us.
Currently our main focus is to leverage ML techniques, in particular deep neural networks, for reconstruction and analysis of data collected by a time projection chamber (TPC), a type of particle detector that allows 2D or 3D imaging of charged particle trajectories. We work on existing TPC detector data as well as DUNE, a future flagship experiment of the U.S. Department of Energy for the next decade. Despite our current focus on TPCs, new contributions to other types of data would be welcomed.
We are keen to share our work in public. You can find our active software development in public github repository, and we make most of our data available in public. We take open-software and open-data effort seriously to make our work as much transparent and reproducible as possible. We also share our experience and interesting findings on our blog post and our weekly group meetings.
Our group consists of physicists at many levels (students, post-docs, lab scientists and faculties) and we are keen to share development effort in all aspects of the research, from software tools and algorithms to high level physics analysis methods.
