Key publications
1. Laurits Tani, Joosep Pata, and Joschka Birk "Reconstructing hadronically decaying tau leptons with a jet foundation model" In: Submitted to SciPost Core , arXiv: arXiv:2503.19165
Contribution: As the main driver of the analysis, I am the first and corresponding author. I was also responsible for producing training samples and data management. I implemented ML models for the studies, developed the code base, analyzed the results, prepared the figures, and contributed substantially to writing the manuscript.
2. Laurits Tani et al. "A unified machine learning approach for reconstructing hadronically decaying tau leptons" In: Comput. Phys. Commun. 307 (Feb. 2025), p. 109399. i, DOI: 10.1016/j.cpc.2024. 109399 , arXiv: 2407.06788 [hep-ex]
Contribution: As the main driver of the analysis, I was the first and corresponding author. I was also responsible for producing training samples and data management. I implemented ML models for the studies, developed the code base, analyzed the results, prepared the figures, and contributed substantially to writing the manuscript. As the corresponding author I also handled the peer-review. Presented the results in a conference.
3. Laurits Tani "Measurement of Higgs Boson Properties in Leptonic Final States using ML-methods" In: PhD thesis. Tallinn U. Tech., 2024. i, DOI: 10.23658/taltech.25/2024
Contribution: PhD dissertation
4. Torben Lange et al. "Tau lepton identification and reconstruction: A new frontier for jet-tagging ML algorithms" In: Comput. Phys. Commun. 298 (2024), p. 109095 i, DOI: 10.1016/j.cpc.2024.109095 , arXiv: 2201.06809 [physics.data-an]
Contribution: I was responsible for producing the samples for the training and developing the code base for the project. I analyzed the results, prepared the figures and contributed substantially to writing the manuscript. As the main driver of the analysis, I was the corresponding author, and was responsible for the peer-review process. Presented the results in a conference.
5. Laurits Tani and Christian Veelken "Comparison of Bayesian and particle swarm algorithms for hyperparameter optimisation in machine learning applications in high energy physics" In: Comput. Phys. Commun. 294 (2024), p. 108955 i, DOI: 10.1016/j.cpc.2023.108955 , arXiv: 2206.10268 [hep-ex]
Contribution: I implemented the algorithms from scratch, and utilized the evaluation framework I developed for the previous hyperparameter optimization paper. As the lead author, I prepared all experimental results and figures, and wrote the manuscript. As the main driver of the analysis, I was the first and corresponding author, handling peer-review. Also, presented the results in a conference.
6. CMS Collaboration "Search for Higgs boson pairs decaying to WW∗WW∗, WW∗ττ, and ττττ in proton-proton collisions at √s = 13 TeV" In: JHEP 07 (2023), p. 095 i, DOI: 10.1007/JHEP07(2023)095 , arXiv: 2206.10268 [hep-ex]
Contribution: I was responsible for the two analysis channels targeting the 4τ decay mode. Additionally I created the ML framework to to be used across the analysis. Wrote parts of the analysis note and contributed in reviewing the final paper.
7. Laurits Tani et al. "Evolutionary algorithms for hyperparameter optimization in machine learning for application in high energy physics" In: Eur. Phys. J. C 81.2 (2021), p. 170 i, DOI: 10.1140/epjc/s10052-021-08950- y , arXiv: 2011.04434 [hep-ex]
Contribution: I implemented the algorithms from scratch and developed the evaluation framework. As the lead author, I prepared all experimental results and figures, and wrote the manuscript. As the main driver of the analysis, I was the first and corresponding author, handling peer-review. Also, presented the results in a conference.
8. Laurits Tani "Monitoring the optical quality of the FACT Cherenkov Telescope" In: MSc thesis, ETH Zürich, 2019 , arXiv: 2001.06712 [astro-ph.IM]
Contribution: MSc thesis. Wrote the air shower simulation from scratch, developed methods to identify muon rings from gammaray and cosmic ray showers. Quantified the muon ring spread to evaluate the image quality and applied the methods on real data. Results later presented in a conference.
