首 页機構概況機構設置科研人才隊伍合作交流研究生教育博士後圖書館創新文化黨群園地院重點實驗室彭桓武中心信息公開
  學術活動
  您現在的位置:首頁 > 學術活動 > 專題學術報告/Seminar
(Seminar) Hunting “Strange” Signals via Deep Learning
2019-11-19  【 】【打印】【關閉

CAS Key Laboratory of Theoretical Physics

Institute of Theoretical Physics

Chinese Academy of Sciences

 Seminar

Title

題目

Hunting “Strange” Signals via Deep Learning

Speaker

報告人

Yuichiro Nakai

Affiliation

所在單位

Tsung-Dao Lee Institute and Shanghai Jiao Tong University

Date

日期

3:00pm, Nov 19, 2019, Tuesday

Venue

地點

ITP South Building 6420

Abstract

摘要

Deep learning is receiving increased attention throughout physics community as well as the real world. In this talk, after a brief introduction of deep learning, I will present two of my recent research on this technique applied to collider physics. The first part of the talk is on the possibility of strange-quark tagging, the last missing piece among quark and gluon identifications in jets. I will describe how to overcome the most difficult classification between strange and down quark jets. Neural networks feed jet images and learn features of strange jets in a supervised way. The second part is on an unsupervised learning technique called autoencoder as a tool for new physics search. The key idea of the autoencoder is that it learns to map background events back to themselves, but fails to reconstruct anomalous events that it has never encountered before. The reconstruction error can then be used as an anomaly threshold. As the first baby step, the example of finding top and gluino jets from background QCD jets will be discussed.

Contact Person

所內聯系人

舒菁
IE6.0浏览器,1024X768分辨率 版权所有 ? 中国科学院理论物理研究所
地址:北京市海澱區中關村東路55號 郵政編碼:100190
京ICP備05002865號】 京公网安备1101080094号