Deep Learning, Dark Knowledge, and Dark Matter

Peter Sadowski, Julian Collado, Daniel Whiteson, Pierre Baldi
Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning, PMLR 42:81-87, 2015.

Abstract

Particle colliders are the primary experimental instruments of high-energy physics. By creating conditions that have not occurred naturally since the Big Bang, collider experiments aim to probe the most fundamental properties of matter and the universe. These costly experiments generate very large amounts of noisy data, creating important challenges and opportunities for machine learning. In this work we use \emphdeep learning to greatly improve the statistical power on three benchmark problems involving: (1) Higgs bosons; (2) supersymmetric particles; and (3) Higgs boson decay modes. This approach increases the expected discovery significance over traditional shallow methods, by 50%, 2%, and 11% respectively. In addition, we explore the use of model compression to transfer information (\emphdark knowledge) from deep networks to shallow networks.

Cite this Paper


BibTeX
@InProceedings{pmlr-v42-sado14, title = {Deep Learning, Dark Knowledge, and Dark Matter}, author = {Sadowski, Peter and Collado, Julian and Whiteson, Daniel and Baldi, Pierre}, booktitle = {Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning}, pages = {81--87}, year = {2015}, editor = {Cowan, Glen and Germain, Cécile and Guyon, Isabelle and Kégl, Balázs and Rousseau, David}, volume = {42}, series = {Proceedings of Machine Learning Research}, address = {Montreal, Canada}, month = {13 Dec}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v42/sado14.pdf}, url = {https://proceedings.mlr.press/v42/sado14.html}, abstract = {Particle colliders are the primary experimental instruments of high-energy physics. By creating conditions that have not occurred naturally since the Big Bang, collider experiments aim to probe the most fundamental properties of matter and the universe. These costly experiments generate very large amounts of noisy data, creating important challenges and opportunities for machine learning. In this work we use \emphdeep learning to greatly improve the statistical power on three benchmark problems involving: (1) Higgs bosons; (2) supersymmetric particles; and (3) Higgs boson decay modes. This approach increases the expected discovery significance over traditional shallow methods, by 50%, 2%, and 11% respectively. In addition, we explore the use of model compression to transfer information (\emphdark knowledge) from deep networks to shallow networks.} }
Endnote
%0 Conference Paper %T Deep Learning, Dark Knowledge, and Dark Matter %A Peter Sadowski %A Julian Collado %A Daniel Whiteson %A Pierre Baldi %B Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Glen Cowan %E Cécile Germain %E Isabelle Guyon %E Balázs Kégl %E David Rousseau %F pmlr-v42-sado14 %I PMLR %P 81--87 %U https://proceedings.mlr.press/v42/sado14.html %V 42 %X Particle colliders are the primary experimental instruments of high-energy physics. By creating conditions that have not occurred naturally since the Big Bang, collider experiments aim to probe the most fundamental properties of matter and the universe. These costly experiments generate very large amounts of noisy data, creating important challenges and opportunities for machine learning. In this work we use \emphdeep learning to greatly improve the statistical power on three benchmark problems involving: (1) Higgs bosons; (2) supersymmetric particles; and (3) Higgs boson decay modes. This approach increases the expected discovery significance over traditional shallow methods, by 50%, 2%, and 11% respectively. In addition, we explore the use of model compression to transfer information (\emphdark knowledge) from deep networks to shallow networks.
RIS
TY - CPAPER TI - Deep Learning, Dark Knowledge, and Dark Matter AU - Peter Sadowski AU - Julian Collado AU - Daniel Whiteson AU - Pierre Baldi BT - Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning DA - 2015/08/27 ED - Glen Cowan ED - Cécile Germain ED - Isabelle Guyon ED - Balázs Kégl ED - David Rousseau ID - pmlr-v42-sado14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 42 SP - 81 EP - 87 L1 - http://proceedings.mlr.press/v42/sado14.pdf UR - https://proceedings.mlr.press/v42/sado14.html AB - Particle colliders are the primary experimental instruments of high-energy physics. By creating conditions that have not occurred naturally since the Big Bang, collider experiments aim to probe the most fundamental properties of matter and the universe. These costly experiments generate very large amounts of noisy data, creating important challenges and opportunities for machine learning. In this work we use \emphdeep learning to greatly improve the statistical power on three benchmark problems involving: (1) Higgs bosons; (2) supersymmetric particles; and (3) Higgs boson decay modes. This approach increases the expected discovery significance over traditional shallow methods, by 50%, 2%, and 11% respectively. In addition, we explore the use of model compression to transfer information (\emphdark knowledge) from deep networks to shallow networks. ER -
APA
Sadowski, P., Collado, J., Whiteson, D. & Baldi, P.. (2015). Deep Learning, Dark Knowledge, and Dark Matter. Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning, in Proceedings of Machine Learning Research 42:81-87 Available from https://proceedings.mlr.press/v42/sado14.html.

Related Material