Accelerating high-energy physics exploration with deep learning

In this work, we present our approach to using deep learning for identification of rarely produced physics particles (such as the Higgs Boson) out of a majority of uninteresting, background or noise-dominated data. A fast and efficient system to eliminate uninteresting data would result in much less data being stored, thus significantly reducing processing time and storage requirements. In this paper, we present a generalized preliminary version of our approach to motivate research interest in advancing the state-of-the-art in deep learning networks for other applications that can benefit from learning systems.