BitHEP -- The Limits of Low-Precision ML in HEP
This work addresses computational bottlenecks in HEP by assessing low-precision ML, but it is incremental as it tests an existing method on new data without major breakthroughs.
The paper evaluated the BitNet architecture for high-energy physics tasks like quark-gluon discrimination and detector simulation, finding it competitive in classification but with variable performance in regression and generation depending on network specifics.
The increasing complexity of modern neural network architectures demands fast and memory-efficient implementations to mitigate computational bottlenecks. In this work, we evaluate the recently proposed BitNet architecture in HEP applications, assessing its performance in classification, regression, and generative modeling tasks. Specifically, we investigate its suitability for quark-gluon discrimination, SMEFT parameter estimation, and detector simulation, comparing its efficiency and accuracy to state-of-the-art methods. Our results show that while BitNet consistently performs competitively in classification tasks, its performance in regression and generation varies with the size and type of the network, highlighting key limitations and potential areas for improvement.