Adaptive Quantization for Hashing: an Information-Based Approach to Learning Binary Codes.
Caiming Xiong,Wei Chen,Gang Chen,David M. Johnson,Jason J. Corso
DOI: https://doi.org/10.1137/1.9781611973440.20
2014-01-01
Abstract:Previous chapter Next chapter Full AccessProceedings Proceedings of the 2014 SIAM International Conference on Data Mining (SDM)Adaptive Quantization for Hashing: An Information-Based Approach to Learning Binary CodesCaiming Xiong, Wei Chen, Gang Chen, David Johnson, and Jason J. CorsoCaiming XiongDepartment of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Search for more papers by this author, Wei ChenDepartment of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Search for more papers by this author, Gang ChenDepartment of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Search for more papers by this author, David JohnsonDepartment of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Search for more papers by this author, and Jason J. CorsoDepartment of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Department of Computer Science and Engineering, SUNY at Buffalo.Search for more papers by this authorpp.172 - 180Chapter DOI:https://doi.org/10.1137/1.9781611973440.20PDFBibTexSections ToolsAdd to favoritesDownload CitationsTrack CitationsEmail SectionsAboutAbstract Large-scale data mining and retrieval applications have increasingly turned to compact binary data representations as a way to achieve both fast queries and efficient data storage; many algorithms have been proposed for learning effective binary encodings. Most of these algorithms focus on learning a set of projection hyperplanes for the data and simply binarizing the result from each hyperplane, but this neglects the fact that informativeness may not be uniformly distributed across the projections. In this paper, we address this issue by proposing a novel adaptive quantization (AQ) strategy that adaptively assigns varying numbers of bits to different hyperplanes based on their information content. Our method provides an information-based schema that preserves the neighborhood structure of data points, and we jointly find the globally optimal bit-allocation for all hyperplanes. In our experiments, we compare with state-of-the-art methods on four large-scale datasets and find that our adaptive quantization approach significantly improves on traditional hashing methods. Previous chapter Next chapter RelatedDetails Published:2014eISBN:978-1-61197-344-0 https://doi.org/10.1137/1.9781611973440Book Series Name:ProceedingsBook Code:PRDT14Book Pages:1-1086