BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//https://techplay.jp//JP
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALDESC:Quantum Machine Learning Seminar (Dr. Hayata Yamasaki\, IQOQI 
 Vienna\, TU Wien)
X-WR-CALNAME:Quantum Machine Learning Seminar (Dr. Hayata Yamasaki\, IQOQI 
 Vienna\, TU Wien)
X-WR-TIMEZONE:Asia/Tokyo
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
BEGIN:STANDARD
DTSTART:19700101T000000
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:824021@techplay.jp
SUMMARY:Quantum Machine Learning Seminar (Dr. Hayata Yamasaki\, IQOQI Vienn
 a\, TU Wien)
DTSTART;TZID=Asia/Tokyo:20210727T170000
DTEND;TZID=Asia/Tokyo:20210727T183000
DTSTAMP:20260422T145329Z
CREATED:20210707T060017Z
DESCRIPTION:イベント詳細はこちら\nhttps://techplay.jp/event/82402
 1?utm_medium=referral&utm_source=ics&utm_campaign=ics\n\n量子機械学
 習で先駆的な研究をされている山崎隼汰氏に、量子情
 報の基本的な事項から始めて最近の研究内容について
 解説して頂きます。量子情報と機械学習の新たな交流
 が生まれることを期待して、ディスカッションの時間
 を少し長めに設定しています。使用言語は日本語です
 。\n\nSpeaker: Dr. Hayata Yamasaki\n\nAffiliation:\n1. Institute for Qua
 ntum Optics and Quantum Information (IQOQI)\, Austrian Academy of Science
 s\n2. Atominstitut\, Technische Universität Wien\n\nPosition:Postdoc\n\n
 Title:\nLearning with Optimized Random Features: Quantum Computation for 
 Accelerating Machine Learning\n\nAbstract:\nThis talk will review the bas
 ics of quantum computation\, and a series of recent works on quantum mach
 ine learning (QML) with optimized random features. The goal of the talk i
 s to explain how to use exponential speedup achieved by quantum computati
 on to accelerate learning without imposing restrictive assumptions.\nRand
 om features are a central technique for scalable learning algorithms base
 d on kernel methods. A recent work has shown that an algorithm using quan
 tum computation can exponentially speed up sampling of optimized random f
 eatures\, even without imposing restrictive assumptions on sparsity and l
 ow-rankness of matrices that had limited applicability of conventional QM
 L algorithms. This QML algorithm makes it possible to significantly reduc
 e and provably minimize the required number of features for achieving lea
 rning tasks.\nThis talk will present applications of this QML algorithm t
 o significant acceleration of leading regression and classification algor
 ithms based on kernel methods\, based on the following papers.\nhttps://a
 rxiv.org/abs/2004.10756\nhttps://arxiv.org/abs/2106.09028\n\n\n\nAll part
 icipants are required to agree with the AIP Open Seminar Series Code of C
 onduct.\nPlease see the URL below.\nhttps://aip.riken.jp/event-list/terms
 ofparticipation/?lang=ja\nRIKEN AIP will expect adherence to this code th
 roughout the event. We expect cooperation from all participants to help e
 nsure a safe environment for everybody.\n\n
LOCATION:オンライン
URL:https://techplay.jp/event/824021?utm_medium=referral&utm_source=ics&utm
 _campaign=ics
END:VEVENT
END:VCALENDAR