Technical Report of Participation in Higgs Boson Machine Learning Challenge
This is an incremental technical report on applying existing methods to a specific physics competition dataset.
The authors tackled the Higgs Boson Machine Learning Challenge by implementing and comparing various machine learning models, including deep learning techniques built from scratch in Python and NumPy, but no concrete results or numbers are reported.
This report entails the detailed description of the approach and methodologies taken as part of competing in the Higgs Boson Machine Learning Competition hosted by Kaggle Inc. and organized by CERN et al. It briefly describes the theoretical background of the problem and the motivation for taking part in the competition. Furthermore, the various machine learning models and algorithms analyzed and implemented during the 4 month period of participation are discussed and compared. Special attention is paid to the Deep Learning techniques and architectures implemented from scratch using Python and NumPy for this competition.