BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260605T171617EDT-9715Kk5XoO@132.216.98.100
DTSTAMP:20260605T211617Z
DESCRIPTION:Machine learning from weak supervision - Towards accurate class
 ification with low labeling costs.\n\nRecent advances in machine learning 
 with big labeled data allow us to achieve human-level performance in vario
 us tasks such as speech recognition\, image understanding\, and natural la
 nguage translation. On the other hand\, there are still many application d
 omains where human labor is involved in the data acquisition process and t
 hus the use of massive labeled data is prohibited.  In this talk\, I will 
 introduce our recent advances in classification techniques from weak super
 vision\, including classification from positive and unlabeled data\, a nov
 el approach to semi-supervised classification\, classification from positi
 ve-confidence data\, and classification from complementary labels\n
DTSTART:20180426T143000Z
DTEND:20180426T153000Z
LOCATION:Room AA-3195\, CA\, Pav. André-Aisenstadt
SUMMARY:Masashi Sugiyama\, RIKEN and University of Tokyo
URL:/mathstat/channels/event/masashi-sugiyama-riken-an
 d-university-tokyo-286740
END:VEVENT
END:VCALENDAR