HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Conference papers

Neural Associative Memories as Accelerators for Binary Vector Search

Chendi Yu 1 Vincent Gripon 2, 1 Xiaoran Jiang 2, 1 Hervé Jégou 3
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance
3 LinkMedia - Creating and exploiting explicit links between multimedia fragments
Inria Rennes – Bretagne Atlantique , IRISA-D6 - MEDIA ET INTERACTIONS
Abstract : Associative memories aim at matching an input noisy vector with a stored one. The matched vector satisfies a minimum distance criterion with respect to the inner metric of the device. This problem of finding nearest neighbors in terms of Euclidean or Hamming distances is a very common operation in machine learning and pattern recognition. However, the inner metrics of associative memories are often misfitted to handle practical scenarios. In this paper, we adapt Willshaw networks in order to use them for accelerating nearest neighbor search with limited impact on accuracy. We provide a theoretical analysis of our method for binary sparse vectors. We also test our method using the MNIST handwritten digits database. Both our analysis for synthetic data and experiments with real-data evidence a significant gain in complexity with negligible loss in performance compared to exhaustive search.
Complete list of metadata

Contributor : Bibliothèque Télécom Bretagne Connect in order to contact the contributor
Submitted on : Wednesday, May 4, 2016 - 2:28:44 PM
Last modification on : Monday, March 14, 2022 - 11:08:11 AM


  • HAL Id : hal-01311588, version 1


Chendi Yu, Vincent Gripon, Xiaoran Jiang, Hervé Jégou. Neural Associative Memories as Accelerators for Binary Vector Search. COGNITIVE 2015 : 7th International Conference on Advanced Cognitive Technologies and Applications, Mar 2015, Nice, France. pp.85 - 89. ⟨hal-01311588⟩



Record views