BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Inverted Multi-Index - Victor Lempitsky\, Yandex
DTSTART:20120607T130000Z
DTEND:20120607T140000Z
UID:TALK38011@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:I will present a new data structure for efficient similarity s
 earch in very large datasets of high-dimensional vectors. This structure c
 alled the inverted multi-index generalizes the inverted index idea by repl
 acing the standard quantization within inverted indices with product quant
 ization. For very similar retrieval complexity and pre-processing time\, i
 nverted multi-indices achieve a much denser subdivision of the search spac
 e compared to inverted indices\, while retaining their memory efficiency. 
 Our experiments with large datasets of SIFT and GIST vectors demonstrate t
 hat because of the denser subdivision\, inverted multi-indices are able to
  return much shorter candidate lists with higher recall. Augmented with a 
 suitable reranking procedure\, multi-indices were able to improve the spee
 d of approximate nearest neighbor search on the dataset of 1 billion SIFT 
 vectors by an order of magnitude compared to the best previously published
  systems\, while achieving better recall and incurring only few percent of
  memory overhead. This is a joint work with Artem Babenko.
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR