BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Unsupervised domain adaptation for human activity recognition - Ju
 an Ye (University of St Andrews)
DTSTART:20201123T140000Z
DTEND:20201123T150000Z
UID:TALK151954@talks.cam.ac.uk
CONTACT:Lorena Qendro
DESCRIPTION:*Abstract:*\nDriven by the persuasiveness of sensing technolog
 ies in our everyday devices\, activity recognition has been widely adopted
  in many applications\, from smart home\, personal healthcare\, human-comp
 uter interaction\, to name a few examples. The key enabler to these applic
 ation is the ability to accurately recognise people's activities. Activity
  recognition techniques\, especially those based on deep neural networks\,
  have made significant progress on learning complex correlations between s
 ensor data and activity classes. However\, they often rely on a large amou
 nt of well-annotated training data\, typically time- and effort-expensive\
 , if not feasible at all\, to acquire. To tackle this challenge\, we have 
 designed various unsupervised domain adaptation techniques to enable trans
 ferring the activity knowledge from one environment to many other environm
 ents. In this talk\, we will introduce 3 of our recent developed technique
 s: (1) knowledge-driven ensemble learning\, (2) variational Autoencoder-ba
 sed feature space alignment\, and (3) bi-directional generative adversaria
 l networks-based feature space transfer. \n\n*Bio:* Dr. Juan Ye is a Senio
 r Lecturer in the School of Computer Science at the University of St Andre
 ws. She received a Bachelor's and Master's degree from Wuhan University\, 
 China\, in 2002 and 2005 respectively\, and a PhD in Computer Science from
  University College Dublin\, Ireland\, in 2009. Her research interest cent
 ers on human activity recognition\, with a particular focus on domain adap
 tation\, continual learning\, and unsupervised learning. 
LOCATION:Virtual (see abstract for Zoom link)
END:VEVENT
END:VCALENDAR