BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Meta-Learning for Versatile Artificial Intelligence
DTSTART:20190717T140000
DTEND:20190717T160000
DTSTAMP:20260406T172954Z
UID:329f69ea7534a27494a7417118c8629a95eb01f1d1cd81435707162f
CATEGORIES:Conferences - Seminars
DESCRIPTION:Arnout Devos\nEDIC candidacy exam\nExam president: Prof. Patri
 ck Thiran\nThesis advisor: Prof. Matthias Grossglauser\nCo-examiner: Prof.
  Martin Jaggi\n\nAbstract\nHumans can learn new concepts from only a few e
 xamples and quickly adapt to unforeseen circumstances. To this end\, they 
 build upon their prior experience and prepare for the ability to adapt. Mo
 st machine learning systems\, however\, lack the versatility to learn new 
 tasks efficiently and effectively when little data is available. Learning 
 to learn\, or meta-learning\, is a paradigm which successfully trains mach
 ine learning models for effective adaptation with little new task data. In
  this work we discuss fundamental and recent advances in meta-learning and
  few-shot learning. The model-agnostic meta-learning (MAML) method and the
  differentiable plasticity method are discussed in detail. We review our p
 ublished and ongoing work on meta-learning\, reaching state-of-the-art per
 formance\, and propose possible future research directions.\n\nBackground 
 papers\nA Closer Look at Few-shot Classification\, by Wei-Yu Chen\, Yen-Ch
 eng Liu\, Zsolt Kira\, Yu-Chiang Frank Wang\, Jia-Bin Huang\, ICLR 2019.\n
 Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks\, by Che
 lsea Finn\, Pieter Abbeel\, Sergey Levine\, ICML 2017.\nDifferentiable pla
 sticity: training plastic neural networks with backpropagation\, by Thomas
  Miconi\, Kenneth Stanley\, Jeff Clune\, ICML 2018.\n\n 
LOCATION:BC 129 https://plan.epfl.ch/?room==BC%20129
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR