BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Geordie Williamson (University of Sydney)
DTSTART:20220224T040000Z
DTEND:20220224T060000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/1
DESCRIPTION:by Geordie Williamson (University of Sydney) as part of SMRI S
 eminar Series: Machine learning for the working mathematician\n\nLecture h
 eld in Carslaw 273 & Online.\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Adam Zsolt Wagner (ETH Zurich)
DTSTART:20220407T050000Z
DTEND:20220407T070000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/2
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/Machi
 neLearning/2/">A simple RL setup to find counterexamples to conjectures in
  mathematics</a>\nby Adam Zsolt Wagner (ETH Zurich) as part of SMRI Semina
 r Series: Machine learning for the working mathematician\n\nLecture held i
 n Carslaw 273 & Online.\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Alex Davies
DTSTART:20220505T060000Z
DTEND:20220505T080000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/3
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/Machi
 neLearning/3/">A technical history of AlphaZero</a>\nby Alex Davies as par
 t of SMRI Seminar Series: Machine learning for the working mathematician\n
 \nLecture held in Carslaw 273 & Online.\n\nAbstract\nIn 2016 AlphaGo defea
 ted the world champion go player Lee Sedol in a historic 5 game match. In 
 this lecture we will discuss the research behind this system and the innov
 ations that ultimately lead to AlphaZero\, which can learn to play multipl
 e board games\, including Go\, from scratch without human knowledge.\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Daniel Halpern-Leinster
DTSTART:20220511T230000Z
DTEND:20220512T010000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/4
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/Machi
 neLearning/4/">Learning selection strategies in Buchberger's algorithm</a>
 \nby Daniel Halpern-Leinster as part of SMRI Seminar Series: Machine learn
 ing for the working mathematician\n\nLecture held in Carslaw 273 & Online.
 \nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lars Buesing
DTSTART:20220519T060000Z
DTEND:20220519T080000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/5
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/Machi
 neLearning/5/">Searching for Formulas and Algorithms: Symbolic Regression 
 and Program Induction</a>\nby Lars Buesing as part of SMRI Seminar Series:
  Machine learning for the working mathematician\n\nLecture held in Carslaw
  273 & Online.\n\nAbstract\nIn spite of their enormous success as black bo
 x function approximators in many fields such as computer vision\, natural 
 language processing and automated decision making\, Deep Neural Networks o
 ften fall short of providing interpretable models of data. In applications
  where aiding human understanding is the main goal\, describing regulariti
 es in data with compact formuli promises improved interpretability and bet
 ter generalization. In this talk I will introduce the resulting problem of
  Symbolic Regression and its generalization to Program Induction\, highlig
 ht some learning methods from the literature and discuss challenges and li
 mitations of searching for algorithmic descriptions of data.\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Qianxiao Li
DTSTART:20220526T050000Z
DTEND:20220526T070000Z
DTSTAMP:20260424T221403Z
UID:MachineLearning/6
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/Machi
 neLearning/6/">Deep learning for sequence modelling</a>\nby Qianxiao Li as
  part of SMRI Seminar Series: Machine learning for the working mathematici
 an\n\nLecture held in Carslaw 273 & Online.\n\nAbstract\nIn this talk\, we
  introduce some deep learning based approaches for modelling sequence to s
 equence relationships that are gaining popularity in many applied fields\,
  such as time-series analysis\, natural language processing\, and data-dri
 ven science and engineering. We will also discuss some interesting mathema
 tical issues underlying these methodologies\, including approximation theo
 ry and optimization dynamics.\n
LOCATION:https://stable.researchseminars.org/talk/MachineLearning/6/
END:VEVENT
END:VCALENDAR