BEGIN:VCALENDAR
PRODID:-//Mozilla.org/NONSGML Mozilla Calendar V1.1//EN
VERSION:2.0
METHOD:REPLY
BEGIN:VTIMEZONE
TZID:Europe/Vienna
X-TZINFO:Europe/Vienna[2023c]
BEGIN:STANDARD
DTSTART:18930401T000000
RDATE:18930401T000000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+010521
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19161001T010000
RDATE:19161001T010000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19170917T030000
RRULE:FREQ=YEARLY;BYMONTH=9;BYDAY=3MO;UNTIL=19180916T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19200913T030000
RDATE:19200913T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19421102T030000
RDATE:19421102T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19431004T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1MO;UNTIL=19441002T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19450412T030000
RDATE:19450412T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19461007T030000
RDATE:19461007T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19471005T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU;UNTIL=19481003T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19800928T000000
RDATE:19800928T000000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19810927T030000
RRULE:FREQ=YEARLY;BYMONTH=9;BYDAY=-1SU;UNTIL=19950924T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19961027T030000
RDATE:19961027T030000
TZNAME:Europe/Vienna(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:STANDARD
DTSTART:19971026T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
TZNAME:(STD)
TZOFFSETFROM:+020000
TZOFFSETTO:+010000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:19160430T230000
RDATE:19160430T230000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19170416T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=3MO;UNTIL=19180415T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19200405T020000
RDATE:19200405T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19400401T020000
RDATE:19400401T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19430329T020000
RDATE:19430329T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19440403T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1MO;UNTIL=19450402T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19460414T020000
RDATE:19460414T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19470406T020000
RDATE:19470406T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19480418T020000
RDATE:19480418T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19800406T000000
RDATE:19800406T000000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19810329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU;UNTIL=19960331T020000
TZNAME:Europe/Vienna(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:19970330T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
TZNAME:(DST)
TZOFFSETFROM:+010000
TZOFFSETTO:+020000
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
CREATED:20231006T120002Z
LAST-MODIFIED:20231006T120108Z
DTSTAMP:20231006T120108Z
UID:3bb18e6f-c914-42ed-b2e5-92580a9a60d8
SUMMARY:Talk by Clemens Heitzinger: Reinforcement Learning and its Applicat
 ion in Medicine and Large Language Models
ORGANIZER;PARTSTAT=NEEDS-ACTION;CUTYPE=INDIVIDUAL:mailto:tristan.miller@ofa
 i.at
ATTENDEE;PARTSTAT=ACCEPTED;CUTYPE=INDIVIDUAL;ROLE=REQ-PARTICIPANT:mailto:tr
 istan.miller@ofai.at
DTSTART;TZID=Europe/Vienna:20240131T183000
DTEND;TZID=Europe/Vienna:20240131T200000
DESCRIPTION;ALTREP="data:text/html,%3Cdiv%20class%3D%22col%20running-text%2
 2%3E%0A%20%20%20%20%20%20%20%20%20%20%3Cp%3EReinforcement%20learning%20has%
 20been%20instrumental%20in%20many%20%0Aadvances%20in%20AI%2C%20including%20
 medicine.%20In%20such%20applications%2C%20statements%20%0Aabout%20the%20rel
 iability%20of%20the%20results%20are%20necessary%20in%20addition%20to%20%0Ac
 onvergence%20results.%20Research%20in%20this%20direction%20is%20the%20topic
 %20of%20%3Cstrong%3E%22Reinforcement%20Learning%20and%20its%20Application%2
 0in%20Medicine%20and%20Large%20Language%20Models%22%3C%2Fstrong%3E%2C%20a%2
 0talk%20by%20%3Cstrong%3EClemens%20Heitzinger%3C%2Fstrong%3E%20of%20%3Cstro
 ng%3ETU%20Wien%3C%2Fstrong%3E.%20%20The%20talk%20is%20part%20of%20%3Ca%20hr
 ef%3D%22https%3A%2F%2Fwww.ofai.at%2Fevents%2Flectures2023fall.html%22%3EOFA
 I's%202023%20Fall%20Lecture%20Series%3C%2Fa%3E.%3C%2Fp%3E%0A%3Cp%3EMembers%
 20of%20the%20public%20are%20cordially%20invited%20to%20attend%20the%20talk%
 20%3Cstrong%3Ein%20person%20(OFAI%2C%20Freyung%206%2F6%2F7%2C%201010%20Vien
 na)%3C%2Fstrong%3E%20or%20via%20Zoom%20on%20%3Cstrong%3EWednesday%2C%2031%2
 0January%202024%20at%2018%3A30%20CET%20(UTC%2B1)%3C%2Fstrong%3E%3A%3C%2Fp%3
 E%0A%3Cp%3EURL%3A%20%3Ca%20href%3D%22https%3A%2F%2Fus06web.zoom.us%2Fj%2F84
 282442460%3Fpwd%3DNHVhQnJXOVdZTWtNcWNRQllaQWFnQT09%22%3Ehttps%3A%2F%2Fus06w
 eb.zoom.us%2Fj%2F84282442460%3Fpwd%3DNHVhQnJXOVdZTWtNcWNRQllaQWFnQT09%3C%2F
 a%3E%3Cbr%3E%0AMeeting%20ID%3A%20842%208244%202460%3Cbr%3E%0APasscode%3A%20
 678868%3C%2Fp%3E%0A%0A%3Cp%3E%3Cstrong%3ETalk%20abstract%3A%3C%2Fstrong%3E%
 20Reinforcement%20learning%20has%20been%20%0Ainstrumental%20in%20many%20adv
 ances%20in%20AI%20in%20recent%20years.%20The%20most%20publicized%0A%20is%20
 certainly%20the%20development%20of%20ChatGPT%20and%20large%20language%20mod
 els%20(LLM)%0A%20in%20general%3B%20the%20last%20and%20crucial%20training%20
 step%20of%20ChatGPT%20is%20%0Areinforcement%20learning%20with%20human%20fee
 dback%20(RLHF).%20Still%2C%20in%20order%20to%20%0Afully%20solve%20learning%
 20problems%2C%20statements%20about%20the%20reliability%20of%20the%20%0Aresu
 lts%20are%20necessary%20in%20addition%20to%20convergence%20results.%20For%2
 0example%2C%20%0Areliability%20and%20trustworthiness%20of%20AI%20systems%20
 is%20of%20utmost%20importance%20in%0A%20medicine%20and%20other%20safety%20c
 ritical%20areas.%20In%20this%20talk%2C%20%0Areinforcement-learning%20algori
 thms%20for%20training%20LLM%20and%20for%20calculating%20%0Aoptimal%20treatm
 ents%20of%20sepsis%20patients%20are%20described.%20The%20questions%20of%20%
 0Aconvergence%20to%20an%20optimal%20policy%20and%20of%20reliability%20are%2
 0addressed%20by%20PAC%0A%20(probably%20approximately%20correct)%20estimates
 %20and%20other%20approaches%20to%20%0Apolicy%20evaluation.%3C%2Fp%3E%0A%3Cp
 %3E%3Cstrong%3ESpeaker%20biography%3A%3C%2Fstrong%3E%20%3Ca%20href%3D%22htt
 p%3A%2F%2FClemens.Heitzinger.name%22%3EClemens%20Heitzinger%3C%2Fa%3E%0A%20
 is%20Co-Director%20of%20the%20Center%20for%20Artificial%20Intelligence%20an
 d%20Machine%20%0ALearning%20(CAIML)%20at%20TU%20Wien%20and%20Associate%20Pr
 ofessor%20in%20the%20Department%20of%0A%20Computer%20Science%20(Informatics
 )%20at%20TU%20Wien.%20He%20received%20both%20his%20%0Amaster's%20degree%20(
 Dipl.-Ing.)%20in%20applied%20mathematics%20and%20his%20PhD%20degree%20%0A(D
 r.%20techn.)%20in%20technical%20sciences%20with%20highest%20honors%20from%2
 0TU%20Wien.%20He%20%0Awas%20a%20visiting%20researcher%20in%20the%20Departme
 nt%20of%20Mathematics%20and%20%0AStatistics%20at%20Arizona%20State%20Univer
 sity%2C%20a%20research%20associate%20in%20the%20%0ASchool%20of%20Electrical
 %20and%20Computer%20Engineering%20at%20Purdue%20University%2C%20and%20a%0A%
 20senior%20research%20associate%20in%20the%20Department%20of%20Applied%20Ma
 thematics%20and%20%0ATheoretical%20Physics%20(DAMTP)%20at%20Cambridge%20Uni
 versity.%20In%202015%2C%20he%20%0Areturned%20to%20TU%20Wien%20as%20an%20ass
 ociate%20professor.%20He%20is%20also%20Adjunct%20%0AProfessor%20in%20the%20
 School%20of%20Mathematical%20and%20Statistical%20Sciences%20at%20%0AArizona
 %20State%20University.%20He%20was%20awarded%20the%20START%20Prize%2C%20Aust
 ria's%20most%0A%20prestigious%20award%20for%20young%20scientists%2C%20by%20
 the%20Austrian%20Science%20Fund%20%0A(FWF)%20in%202013.%20He%20is%20author%
 20of%20%3Cem%3E%3Ca%20href%3D%22https%3A%2F%2Flink.springer.com%2Fbook%2F10
 .1007%2F978-3-031-16560-3%22%3EAlgorithms%20with%20Julia%3C%2Fa%3E%3C%2Fem%
 3E%0A%20(Springer%2C%202022).%20His%20research%20interests%20are%20reinforc
 ement%20learning%20and%0A%20uncertainty%20quantification%20(in%20particular
 %20Bayesian%20inversion)%20with%20%0Aapplications%20in%20the%20sciences%2C%
 20medicine%2C%20and%20engineering.%3C%2Fp%3E%0A%20%20%20%20%20%20%20%20%3C%
 2Fdiv%3E":\n          \n\nReinforcement learning has been instrumental in m
 any \nadvances in AI\, including medicine. In such applications\, statement
 s \nabout the reliability of the results are necessary in addition to \ncon
 vergence results. Research in this direction is the topic of "Reinforcement
  Learning and its Application in Medicine and Large Language Models"\, a ta
 lk by Clemens Heitzinger of TU Wien.  The talk is part of OFAI's 2023 Fall 
 Lecture Series.\n\nMembers of the public are cordially invited to attend th
 e talk in person (OFAI\, Freyung 6/6/7\, 1010 Vienna) or via Zoom on Wednes
 day\, 31 January 2024 at 18:30 CET (UTC+1):\n\nURL: https://us06web.zoom.us
 /j/84282442460?pwd=NHVhQnJXOVdZTWtNcWNRQllaQWFnQT09\n\nMeeting ID: 842 8244
  2460\n\nPasscode: 678868\n\n\nTalk abstract: Reinforcement learning has be
 en \ninstrumental in many advances in AI in recent years. The most publiciz
 ed\n is certainly the development of ChatGPT and large language models (LLM
 )\n in general\; the last and crucial training step of ChatGPT is \nreinfor
 cement learning with human feedback (RLHF). Still\, in order to \nfully sol
 ve learning problems\, statements about the reliability of the \nresults ar
 e necessary in addition to convergence results. For example\, \nreliability
  and trustworthiness of AI systems is of utmost importance in\n medicine an
 d other safety critical areas. In this talk\, \nreinforcement-learning algo
 rithms for training LLM and for calculating \noptimal treatments of sepsis 
 patients are described. The questions of \nconvergence to an optimal policy
  and of reliability are addressed by PAC\n (probably approximately correct)
  estimates and other approaches to \npolicy evaluation.\n\nSpeaker biograph
 y: Clemens Heitzinger\n is Co-Director of the Center for Artificial Intelli
 gence and Machine \nLearning (CAIML) at TU Wien and Associate Professor in 
 the Department of\n Computer Science (Informatics) at TU Wien. He received 
 both his \nmaster's degree (Dipl.-Ing.) in applied mathematics and his PhD 
 degree \n(Dr. techn.) in technical sciences with highest honors from TU Wie
 n. He \nwas a visiting researcher in the Department of Mathematics and \nSt
 atistics at Arizona State University\, a research associate in the \nSchool
  of Electrical and Computer Engineering at Purdue University\, and a\n seni
 or research associate in the Department of Applied Mathematics and \nTheore
 tical Physics (DAMTP) at Cambridge University. In 2015\, he \nreturned to T
 U Wien as an associate professor. He is also Adjunct \nProfessor in the Sch
 ool of Mathematical and Statistical Sciences at \nArizona State University.
  He was awarded the START Prize\, Austria's most\n prestigious award for yo
 ung scientists\, by the Austrian Science Fund \n(FWF) in 2013. He is author
  of Algorithms with Julia\n (Springer\, 2022). His research interests are r
 einforcement learning and\n uncertainty quantification (in particular Bayes
 ian inversion) with \napplications in the sciences\, medicine\, and enginee
 ring.\n\n        
LOCATION:In person at OFAI (Freyung 6/6/7\, 1010 Vienna\, Austria) or onlin
 e via Zoom: https://us06web.zoom.us/j/84282442460?pwd=NHVhQnJXOVdZTWtNcWNRQ
 llaQWFnQT09
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR