BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20250330T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260424T085443Z
UID:1733135400@ist.ac.at
DTSTART:20241202T113000
DTEND:20241202T123000
DESCRIPTION:Speaker: Jan Peters\nhosted by Christoph Lampert\nAbstract: Abs
 tract:Autonomous robots that can assist humans in situations of daily life
  have been a long standing vision of robotics\, artificial intelligence\, 
 and cognitive sciences. A first step towards this goal is to create robots
  that can learn tasks triggered by environmental context or higher level i
 nstruction. However\, learning techniques have yet to live up to this prom
 ise as only few methods manage to scale to high-dimensional manipulator or
  humanoid robots. In this talk\, we investigate a general framework suitab
 le for learning motor skills in robotics which is based on the principles 
 behind many analytical robotics approaches. To accomplish robot reinforcem
 ent learning learning from just few trials\, the learning system can no lo
 nger explore all learn-able solutions but has to prioritize one solution o
 ver others – independent of the observed data. Such prioritization requi
 res explicit or implicit assumptions\, often called ‘induction biases’
  in machine learning. Extrapolation to new robot learning tasks requires i
 nduction biases deeply rooted in general principles and domain knowledge f
 rom robotics\, physics and control. Empirical evaluations on a several rob
 ot systems illustrate the effectiveness and applicability to learning cont
 rol on an anthropomorphic robot arm. These robot motor skills range from t
 oy examples (e.g.\, paddling a ball\, ball-in-a-cup) to playing robot tabl
 e tennis\, juggling and manipulation of various objects.Bio:Jan Peters is 
 a full professor (W3) for Intelligent Autonomous Systems at the Computer S
 cience Department of the Technische Universitaet Darmstadt since 2011\, an
 d\, at the same time\, he is the dept head of the research department on S
 ystems AI for Robot Learning (SAIROL) at the German Research Center for Ar
 tificial Intelligence (Deutsches Forschungszentrum für Künstliche Intell
 igenz\, DFKI) since 2022. He is also is a founding research faculty member
  of the Hessian Center for Artificial Intelligence. Jan Peters has receive
 d the Dick Volz Best 2007 US PhD Thesis Runner-Up Award\, the Robotics: Sc
 ience & Systems - Early Career Spotlight\, the INNS Young Investigator Awa
 rd\, and the IEEE Robotics & Automation Society's Early Career Award as we
 ll as numerous best paper awards. In 2015\, he received an ERC Starting Gr
 ant and in 2019\, he was appointed IEEE Fellow\, in 2020 ELLIS fellow and 
 in 2021 AAIA fellow.Despite being a faculty member at TU Darmstadt only si
 nce 2011\, Jan Peters has already nurtured a series of outstanding young r
 esearchers into successful careers. These include new faculty members at l
 eading universities in the USA\, Japan\, Germany\, Finland and Holland\, p
 ostdoctoral scholars at top computer science departments (including MIT\, 
 CMU\, and Berkeley) and young leaders at top AI companies (including Amazo
 n\, Boston Dynamics\, Google and Facebook/Meta).Jan Peters has studied Com
 puter Science\, Electrical\, Mechanical and Control Engineering at TU Muni
 ch and FernUni Hagen in Germany\, at the National University of Singapore 
 (NUS) and the University of Southern California (USC). He has received fou
 r Master's degrees in these disciplines as well as a Computer Science PhD 
 from USC. Jan Peters has performed research in Germany at DLR\, TU Munich 
 and the Max Planck Institute for Biological Cybernetics (in addition to th
 e institutions above)\, in Japan at the Advanced Telecommunication Researc
 h Center (ATR)\, at USC and at both NUS and Siemens Advanced Engineering i
 n Singapore. He has led research groups on Machine Learning for Robotics a
 t the Max Planck Institutes for Biological Cybernetics (2007-2010) and Int
 elligent Systems (2010-2021).
LOCATION:Raiffeisen Lecture Hall\, ISTA
ORGANIZER:maria.arias.sutil@ista.ac.at
SUMMARY:Jan Peters: Inductive Biases for Robot Learning
URL:https://talks-calendar.ista.ac.at/events/5100
END:VEVENT
END:VCALENDAR