Aspects of the Binary CMAC: Unimodularity and Probabilistic Reconstruction

被引：0
作者：
Anil Menon
机构：
[1] ProductSoft Technologies,
来源：
Neural Processing Letters | 2005年 / 22卷
关键词：
CMAC; consecutive ones; interpolation; reconstruction; total unimodularity;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
The CMAC is a neural net for computing real-valued functions. Conceptually, the CMAC maps a point in the function’s domain to a set of locations in an associative memory. Each location “contains” a weight, and the sum of the weights is taken to be the function’s value at that point. The overall process may be modeled as the multiplication of an input vector by an “association matrix.” This paper highlights some aspects of the CMAC’s mapping and function computation procedures. Regarding the mapping procedure, it is shown that the associative matrix of a univariate CMAC has the consecutive-ones property; this implies that the matrix is totally-unimodular, a property of great importance in integer optimization. For a multivariate CMAC, the association matrix can be partitioned into sub-matrices, each with the consecutive-ones property. Regarding the function computation procedure, it is shown that a univariate CMAC can compute a function exactly iff a certain distribution is reconstructible from its one dimensional marginals. A CMAC extension, free of this limitation and derived from the theory of balanced matrices, is briefly discussed. These results are generalizable to multivariate CMACs in a relatively straightforward manner.
引用
收藏
页码：263 / 276
页数：13
相关论文
共 36 条
  • [11] Zabell S.(1997)On the existence of sequences and matrices with prescribed partial sums of elements Linear Algebra And Its Applications 265 71-92
  • [12] Deming F.F.(1996)Hierarchical image coding via cerebellar model arithmetic computers IEEE Transactions On Image Processing 5 1393-1401
  • [13] Stephan N.E.(2000)Patterns that allow give row and column sums Linear Algebra And Its Applications 311 97-105
  • [14] Fulkerson D.R.(1999)Basis function models of the cmac network Neural Networks 12 107-126
  • [15] Gross O.A.(1990)CMAC: An associative neural network alternative to backpropagation Proceedings of the IEEE 78 1561-1567
  • [16] Gonźalez-Serrano F.J.(1989)Convergence properties of the associative memory storage for learning control systems Automation and Remote Control 50 254-286
  • [17] Figueiras–Vidal A.R.(1992)Learning convergence in the cerebellar model articulation controller IEEE Trans. Neural Networks 3 115-121
  • [18] Artés–Rodriguez A.(undefined)undefined undefined undefined undefined-undefined
  • [19] Handelman D.A.(undefined)undefined undefined undefined undefined-undefined
  • [20] Lane S.H.(undefined)undefined undefined undefined undefined-undefined
1234