BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Steps towards making machine learning more natural
DTSTART:20210419T130000
DTEND:20210419T140000
DTSTAMP:20260407T043204Z
UID:c870bea3a0c79b5770d8974a45aec1ab587c78876bcd3b294a6923c7
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Mengye Ren - Uber ATG and University of Toronto\nIC Facult
 y candidate\n\nAbstract\nOver the past decades\, we have seen machine lear
 ning making great strides in AI applications. Yet\, most of its success re
 lies on training models offline on a massive amount of data and evaluating
  them in a similar test environment. By contrast\, humans can learn new co
 ncepts and skills with very few examples\, and can easily generalize to no
 vel tasks. In this talk\, I will highlight three key steps towards making 
 machines learning more human-like\, and these steps will unlock the next g
 eneration of technologies. The first step is to make machines learn new co
 ncepts continually and incrementally using limited labeled data. The secon
 d step is to develop flexible representations that can generalize well to 
 novel concepts under different contexts. Finally\, I’ll show how to make
  abstract and compositional reasoning given visual inputs. I’ll then con
 clude with an outlook of future directions towards building a more general
  and flexible AI.\n\nBio\nMengye Ren is a PhD student in the machine learn
 ing group of the Department of Computer Science at the University of Toron
 to. He was also a research scientist at Uber ATG working on self-driving c
 ars from 2017 to 2021. His research focuses on making machines learn in mo
 re naturalistic environments with less labeled data. He has won a number o
 f awards including two NVIDIA research pioneer awards and the Alexander Gr
 aham Bell Canada Graduate Fellowship.\n\nMore information
LOCATION:https://epfl.zoom.us/j/88338291926?pwd=NFJwWnlNRzI0NEFVM0hzYUFLOW
 NpUT09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR