The Machine Learning landscape of top taggers
The Machine Learning landscape of top taggers
复制标题
DOI：
10.21468/scipostphys.7.1.014
复制
发表时间：
2019-02
期刊：
影响因子：
5.5
通讯作者：
Gregor Kasieczka;T. Plehn;A. Butter;K. Cranmer;Dipsikha Debnath;B. Dillon;M. Fairbairn;D. Faroughy;W. Fedorko;L. Gouskos;J. Kamenik;Patrick T. Komiske;Simon Leiss;A. Lister;S. Macaluso;S. Macaluso;E. Metodiev;L. Moore;B. Nachman;B. Nachman;Karl Nordström;J. Pearkes;H. Qu;Y. Rath;M. Rieger;D. Shih;J. Thompson;Sreedevi Varma
中科院分区：
文献类型：
--
作者：
Gregor Kasieczka;T. Plehn;A. Butter;K. Cranmer;Dipsikha Debnath;B. Dillon;M. Fairbairn;D. Faroughy;W. Fedorko;L. Gouskos;J. Kamenik;Patrick T. Komiske;Simon Leiss;A. Lister;S. Macaluso;S. Macaluso;E. Metodiev;L. Moore;B. Nachman;B. Nachman;Karl Nordström;J. Pearkes;H. Qu;Y. Rath;M. Rieger;D. Shih;J. Thompson;Sreedevi Varma
基于确定增强，强调腐烂的顶级夸克的既定任务，我们比较了广泛的现代机器学习方法。与大多数已建立的方法不同，它们依赖于低级输入，例如热量计输出。尽管他们的网络体系结构大不相同，但它们的性能相对相似。总的来说，我们发现这些新方法非常强大且有趣。
Based on the established task of identifying boosted, hadronically decaying top quarks, we compare a wide range of modern machine learning approaches. Unlike most established methods they rely on low-level input, for instance calorimeter output. While their network architectures are vastly different, their performance is comparatively similar. In general, we find that these new approaches are extremely powerful and great fun.