Multi-Scale Distributed Representation for Deep Learning and its Application to b-Jet Tagging
This work addresses performance bottlenecks in b-jet tagging for high-energy physics, representing an incremental improvement in preprocessing methods.
The paper tackles the problem of improving deep neural network performance for specific high-energy physics tasks by introducing a multi-scale distributed binary representation as a preprocessing step, resulting in significant performance improvements without additional feature engineering.
Recently machine learning algorithms based on deep layered artificial neural networks (DNNs) have been applied to a wide variety of high energy physics problems such as jet tagging or event classification. We explore a simple but effective preprocessing step which transforms each real-valued observational quantity or input feature into a binary number with a fixed number of digits. Each binary digit represents the quantity or magnitude in different scales. We have shown that this approach improves the performance of DNNs significantly for some specific tasks without any further complication in feature engineering. We apply this multi-scale distributed binary representation to deep learning on b-jet tagging using daughter particles' momenta and vertex information.