BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Sabre//Sabre VObject 4.5.8//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Zurich
X-LIC-LOCATION:Europe/Zurich
TZURL:http://tzurl.org/zoneinfo/Europe/Zurich
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19810329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19961027T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:news346@dmi.unibas.ch
DTSTAMP;TZID=Europe/Zurich:20181031T163430
DTSTART;TZID=Europe/Zurich:20181108T161500
SUMMARY:Computer Science Kolloquium: Dr. Evangelos Pournaras\, ETH Zürich
DESCRIPTION:Abstract:\\r\\nThe Internet of Things equips citizens with phen
 omenal new means for online participation in sharing economies. When agent
 s self-determine options from which they choose\, for instance their resou
 rce consumption and production\, while these choices have a collective sys
 tem-wide impact\, optimal decision-making turns into a combinatorial optim
 ization problem known as NP-hard. In such challenging computational proble
 ms\, centrally managed (deep) learning systems often require personal data
  with implications on privacy and citizens’ autonomy. This paper envisio
 ns an alternative unsupervised and decentralized collective learning appro
 ach that preserves privacy\, autonomy and participation of multi-agent sys
 tems self-organized into a hierarchical tree structure. \\r\\nRemote inter
 actions orchestrate a highly efficient process for decentralized collectiv
 e learning. This disruptive concept is realized by I-EPOS\, the Iterative 
 Economic Planning and Optimized Selections\, accompanied by a paradigmatic
  software artifact. Strikingly\, I-EPOS outperforms related algorithms tha
 t involve non-local brute-force operations or exchange full information. T
 his paper contributes new experimental  findings about the influence of n
 etwork topology and planning on learning efficiency as well as  findings 
 on techno-socio-economic trade-offs and global optimality. Experimental ev
 aluation with real-world data from energy and bike sharing pilots demonstr
 ates the grand potential of collective learning to design ethically and so
 cially responsible participatory sharing economies.
X-ALT-DESC:\nAbstract:\nThe Internet of Things equips citizens with phenome
 nal new means for online participation in sharing economies. When agents s
 elf-determine options from which they choose\, for instance their resource
  consumption and production\, while these choices have a collective system
 -wide impact\, optimal decision-making turns into a combinatorial optimiza
 tion problem known as NP-hard. In such challenging computational problems\
 , centrally managed (deep) learning systems often require personal data wi
 th implications on privacy and citizens’ autonomy. This paper envisions 
 an alternative unsupervised and decentralized collective learning approach
  that preserves privacy\, autonomy and participation of multi-agent system
 s self-organized into a hierarchical tree structure. \nRemote interactions
  orchestrate a highly efficient process for decentralized collective learn
 ing. This disruptive concept is realized by I-EPOS\, the Iterative Economi
 c Planning and Optimized Selections\, accompanied by a paradigmatic softwa
 re artifact. Strikingly\, I-EPOS outperforms related algorithms that invol
 ve non-local brute-force operations or exchange full information. This pap
 er contributes new experimental &nbsp\;findings about the influence of net
 work topology and planning on learning efficiency as well as &nbsp\;findin
 gs on techno-socio-economic trade-offs and global optimality. Experimental
  evaluation with real-world data from energy and bike sharing pilots demon
 strates the grand potential of collective learning to design ethically and
  socially responsible participatory sharing economies.\n\n
DTEND;TZID=Europe/Zurich:20181108T180000
END:VEVENT
END:VCALENDAR