BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Convolutional Neural Networks - Christof Angermueller\, Alex Kenda
 ll
DTSTART:20150521T140000Z
DTEND:20150521T153000Z
UID:TALK58587@talks.cam.ac.uk
CONTACT:Rowan McAllister
DESCRIPTION:*Abstract*:\nIn this talk we outline convolutional neural netw
 orks (convnets) and discuss their contemporary applications and research. 
 We begin by outlining prior assumptions and learning techniques for unders
 tanding data with a spatial structure. Secondly\, we go through the key in
 sights that have allowed convnets to surpass state-of-the-art performance 
 in visual classification\, regression\, OCR\, scene understanding and visu
 al reinforcement learning.\n\n*Reading*\n# Jarrett et al.\, "What Is the B
 est Multi-Stage Architecture for Object Recognition?":http://ieeexplore.ie
 ee.org/stamp/stamp.jsp?tp=&arnumber=5459469\n# Krizhevsky\, Sutskever\, an
 d Hinton\, "ImageNet Classification with Deep Convolutional Neural Network
 s.":http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-con
 volutional-neural-networks.pdf\n# Szegedy\, et al. "Going deeper with conv
 olutions.":http://arxiv.org/pdf/1409.4842v1.pdf
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR