摘要: This article proposes to use a binary random matrix with the elements {0,1} project input floating-point vectors onto output of smaller dimension. The accuracies estimates scalar product, Euclidean distance, and norm are analyzed respect vectors. It is analytically experimentally shown that an error for proposed projection than ternary matrix.
springer.com
acm.org
sci-hub.se 参考文章(29)
Oleksiy Dekhtyarenko, Dimitri Nowicki, Averaging on Riemannian Manifolds and Unsupervised Learning using Neural Associative Memory the european symposium on artificial neural networks. pp. 181- 186 ,(2005)
Artem Sokolov, LIMSI: Learning Semantic Similarity by Selecting Random Word Subsets joint conference on lexical and computational semantics. pp. 543- 546 ,(2012)
Alexander A Letichevsky, Olexandr A Letychevskyi, Vladimir S Peschanenko, None, Insertion Modeling System Perspectives of Systems Informatics. pp. 262- 273 ,(2012) , 10.1007/978-3-642-29709-0_23
Alexander Frolov, Dmitri Rachkovskij, Dusan Husek, On information characteristics of sparsely encoded binary auto-associative memory international conference on neural information processing. ,vol. 1, pp. 235- 238 ,(2002) , 10.1109/ICONIP.2002.1202168
Santosh Vempala, The random projection method American Mathematical Society. ,(2005) , 10.1090/DIMACS/065
Jan Kristoferson, Magnus Sahlgren, Jussi Karlgren, Pentti Kanerva, R. Karlsson, Björn Levin, Anders Holst, Gunnar Sjödin, Computing with large random patterns CSLI Publications. pp. 251- 311 ,(2001)
Dmitri A. Rachkovskij, Some approaches to analogical mapping with structure-sensitive distributed representations Journal of Experimental and Theoretical Artificial Intelligence. ,vol. 16, pp. 125- 144 ,(2004) , 10.1080/09528130410001712862
A Frolov, A Kartashov, A Goltsev, R Folk, Quality and efficiency of retrieval for Willshaw-like autoassociative networks. II. Recognition Network: Computation In Neural Systems. ,vol. 6, pp. 513- 534 ,(1995) , 10.1088/0954-898X_6_4_001
D. A. Rachkovskij, E. G. Revunova, A randomized method for solving discrete ill-posed problems Cybernetics and Systems Analysis. ,vol. 48, pp. 621- 635 ,(2012) , 10.1007/S10559-012-9443-6
Alexander A. Letichevsky, Theory of interaction, insertion modeling, and cognitive architectures biologically inspired cognitive architectures. ,vol. 8, pp. 19- 32 ,(2014) , 10.1016/J.BICA.2014.03.001