BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning with Memory Embeddings - Professor Volker Tresp (Ludwig M
 aximilian University of Munich)
DTSTART:20160908T100000Z
DTEND:20160908T110000Z
UID:TALK66812@talks.cam.ac.uk
CONTACT:Louise Segar
DESCRIPTION:Embedding learning\, a.k.a. representation learning\,  has bee
 n shown to be able to model large-scale semantic knowledge graphs. A  key 
 concept is a mapping of the knowledge graph to  a tensor representation wh
 ose entries are predicted by models using latent representations of genera
 lized entities.  Latent variable models are well suited to deal with the h
 igh dimensionality and sparsity of typical knowledge graphs and have succe
 ssfully been employed in knowledge graph completion and fact extraction fr
 om the Web.  We have extended the approach to also consider temporal evolu
 tions\, temporal patterns and subsymbolic representations\, which permits 
 us to model medical decision processes. In addition\, we consider embeddin
 g approaches to be a possible basis for modeling cognitive memory function
 s\, in particular semantic and concept memory\, episodic memory\, sensory 
 memory\, short-term memory\, and working memory.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR