Recent Writing
Identification of highly boosted decays with the ATLAS detector using deep neural networks
This thesis introduces two jet tagging algorithms to identify highly boosted decays using the ATLAS detector at the LHC. Based on the Deep Neural Network (DNN) architecture, the first algorithm's performance is comparable to an existing algorithm designed for highly boosted decays. The DNN jet tagger is also multifunctional and highly effective for identifying decays. Notably, it displayed enhanced rejection rates for background -jets. The second algorithm leverages an Adversarial Neural Network (ANN) architecture for mass-decorrelated classification. While it exhibited a slight performance decrease compared to the DNN-based tagger, it demonstrated a reduction in mutual information between the mass feature and scalar discriminant metric, substantiating its capability for mass-decorrelated jet identification.
Why Code Generating Models Got Good
A literature review examining how code-generating language models improved as their architectures captured more context. We see that a grounded theoretical basis developed from natural language processing spurred the application of increasingly complex statisical methods to this problem.
About
I am a data scientist and machine learning engineer working on the data processing and robotics for IslandView. Previously I built ML and data pipelines for a UK built-environment scanning platform, and before that I did my thesis on a jet-tagging algorithm for boosted Higgs decays in CERN's ATLAS collaboration.
Get In Touch
Interested in collaborating or have questions about my work? Feel free to reach out.