Professor, Department of Physics & Astronomy
- 410 Nieuwland Science Hall
Notre Dame, IN 46556
- +1 574-631-7322
If Prof. Lannon could have the answer to just one question about the universe, it would be why is the top quark so massive while all the other particles we know about have much smaller masses? In 2012 the CMS and ATLAS experiments, using data from proton collisions generated by the LHC at CERN in Geneva, Switzerland, discovered the Higgs boson, confirming that particle mass does, in fact, arise from interacting with the Higgs field. However, there is currently no theory that explains why the Higgs field would interact so much more strongly with one particle, the top quark, than with all the others. It might just be random, but based on examples from history, like the successes of the periodic table, Prof. Lannon is motivated to search for underlying principles to explain the pattern of observed particle masses. Prof. Lannon does this by using the CMS detector to study collisions involving both top quarks and other particles, like Higgs bosons, Z bosons, or W bosons. Using a theoretical framework known as effective field theory (EFT), Prof. Lannon’s research group searches for unexpected features in the data that would point the way towards an explanation for the top quark’s tremendous mass.
Prof. Lannon has developed a keen interest in applying signal processing and big data analysis techniques to confront the challenges posed by the enormous volume of data generated by CMS. He has explored numerous techniques for enhancing the capacity of CMS to use computational resources to process more data. In particular, Prof. Lannon’s group has developed software that enables the use for CMS data processing of opportunistic computing resources at Notre Dame’s Center for Computing (CRC), routinely reaching a scale of 10,000 CPU cores and occasionally peaking at just over 25,000 CPU cores. Prof. Lannon’s group also explores the application of deep learning techniques to particle physics problems. He is the lead principal investigator for the Cyberinfrastructure to Accelerate Machine Learning (CAML) GPU cluster which benefits machine learning efforts across Notre Dame and nationally via the Open Science Grid (OSG). Prof. Lannon also contributes to the CMS trigger. The trigger is responsible for analyzing CMS data in real time to decide whether a particular collision is worth storing for later data analysis. One aspect of the trigger that has become a particular focus for Prof. Lannon is the track trigger upgrade, which seeks to use custom electronics based on field programmable gate arrays (FPGAs) to reconstruct the trajectories of charged particles in time for the trigger to make use of that information.
Honors and Activities
Rev. Edmund P. Joyce, C.S.C., Awards for Excellence in Undergraduate Teaching, 2020
Former REU Projects involving HEP and Computing
- Deep Learning for Particle Physics: Investigating Neural Network Structure and Hyperparameters: https://youtu.be/AQdUIm0N41s
- Lobster: Harnessing Opportunistic Clusters with a Workflow Management Tool for CMS Data Analysis: https://youtu.be/YlV2nfOx-MQ
- Deep Neural Networks for Reconstructing Particle Collisions: https://www.youtube.com/watch?v=n7ZGIn1H4iY
- Analyzing Resource Metadata from High Throughput Computing in an Opportunistic Environment: https://youtu.be/VE3kPxsgfno
- Dynamic ELK Stack Monitoring for Lobster: https://youtu.be/xVQRGYcZrNM
- Using a High-performance, Column-store Database for Particle Physics Analysis: https://youtu.be/nXqc27SQUVM
- Using Deep Neural Networks to Analyze Collisions in High Energy Physics: https://youtu.be/HVYUhK6iOEg
B.S., St. Norbert College, 1997
Ph.D., University of Illinois, Urbana-Champaign, 2003
J. Stietzel and K. Lannon, “Study of Neural Network Size Requirements for Approximating Functions Relevant to HEP,” EPJ Web Conf., 214 (2019).
A. M. Sirunyan et al. [CMS Collaboration], “Observation of ttH production,” Phys. Rev. Lett. 120, no. 23, 231801 (2018).
A. M. Sirunyan et al. [CMS Collaboration], “Evidence for associated production of a Higgs boson with a top quark pair in final states with electrons, muons, and hadronically decaying tau leptons at sqrt(s) = 13 TeV,” JHEP 1808, 066 (2018).
A. M. Sirunyan et al. [CMS Collaboration], “Measurement of the cross section for top quark pair production in association with a W or Z boson in proton-proton collisions at sqrt(s) = 13 TeV,” JHEP 1808, 011 (2018).
P.Ivie, C.Zheng, K.Lannon and D.Thain, “Ananalysis of reproducibility and non-determinism in HEP software and ROOT data,” J. Phys. Conf. Ser. 898, no. 10, 102007 (2017).
M. Wolf et al., “Scaling up a CMS tier-3 site with campus resources and a 100 Gb/s net- work connection: what could go wrong?,” J. Phys. Conf. Ser. 898, no. 8, 082041 (2017)
M. Wolf et al., “Opportunistic Computing with Lobster: Lessons Learned from Scaling up to 25k Non-Dedicated Cores,” J. Phys. Conf. Ser. 898, no. 5, 052036 (2017).
J. Balcas et al., “CMS Connect,” J. Phys. Conf. Ser. 898, no. 8, 082032 (2017).
E. Bartz et al., “FPGA-Based Tracklet Approach to Level-1 Track Finding at CMS for the HL- LHC,” EPJ Web Conf. 150, 00016 (2017)
V. Khachatryan et al. [CMS Collaboration], “Observation of top quark pairs produced in associate with a vector boson in pp collisions at sqrt(s) = 8 TeV,” JHEP 1601, 096 (2016).