BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Informatik Austria - ECPv5.16.4//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Informatik Austria
X-ORIGINAL-URL:https://www.informatikaustria.at
X-WR-CALDESC:Veranstaltungen für Informatik Austria
BEGIN:VTIMEZONE
TZID:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20180325T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20181028T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Berlin:20180509T150000
DTEND;TZID=Europe/Berlin:20180509T164500
DTSTAMP:20260422T211939
CREATED:20180430T091726Z
LAST-MODIFIED:20180430T091726Z
UID:1655-1525878000-1525884300@www.informatikaustria.at
SUMMARY:Data Analytics Without Seeing the Data
DESCRIPTION:CS-COLLOQUIUM mit Dr. Maximilian Ott\n„Today\, we first need to collect data before we can analyse them. This not only creates privacy concerns but also security risks for the collector. For many use cases\, we really only want the analysis and data collection becomes the necessary evil.“ \nOrt: HS3\, Währinger Straße 29\, 1090 Wien \n» Website zur Veranstaltung \n  \nAbstract\nToday\, we first need to collect data before we can analyse them. This not only creates privacy concerns but also security risks for the collector. For many use cases\, we really only want the analysis and data collection becomes the necessary evil. \nIn this talk we describe some of the fundamental techniques which allow us to calculate with encrypted data\, as well as protocols for distributed analysisand associated security modelsto allow us to give formal guarantees on what every party can or more importantly cannot learn during the computation. We will use some of the standard algorithms\, such as logistic regression\, to highlight the differences to conventional big-data analytics frameworks. \nWe will also describe the architecture and some interesting implementation details of our N1 Analytics Platform which is one of the few emerging industry strength (and hopefully\, also open-source) implementations in this space. We will present some performance and scalability measureswe collected from initial customer trials. Finally\, we will conclude with a discussion on some of the challenges we face in developing distributed machine learning algorithm which are not only high-performing\, but also incorporate their data confidentiality claimsin a more formal manner. \nBio\nDr. Maximilian Ottis a Sr. Principal Engineer at Data61 where he is currently developing a platform for federated machine learning over encrypted data. Before that he was a founder and CTO of Incoming Media\, which combined data science and machine learning to create unique subscriber insights and a superior user experience around mobile video. Before coming to Australia where he held various research leadership positions at NICTA\, he was the founder and CTO of Semandex\, US which develops a range of cutting edge semantic information processing technologies and tools for information assurance professionals. He also held  a Research Professorship at Rutgers University where he was heavily involved in GENI\, the US’s Future Internet initiative and various EU FIRE programs. \nHe obtained his PhD from the University in Tokyo and a Dipl.Ing from the Technical University of Vienna.
URL:https://www.informatikaustria.at/event/data-analytics-without-seeing-the-data/
LOCATION:Fakultät für Informatik | Universität Wien\, Währinger Straße 29\, Wien\, 1090\, Österreich
ORGANIZER;CN="Fakult%C3%A4t%20f%C3%BCr%20Informatik":MAILTO:dekanat.informatik@univie.ac.at
END:VEVENT
END:VCALENDAR