Report presents conditional-probability interpretation of Kanerva's sparse distributed memory (SDM). SDM is conceptual digital electronic memory in which addresses chosen sparsely in space of many dimensions and content of memory at each address positive or negative integer written (incremented or decremented) or read according to special rules. Contents of SDM constitute Monte Carlo approximation to multidimensional conditional-probability integral.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Conditional Probability And The Sparse Distributed Memory


    Contributors:

    Published in:

    Publication date :

    1991-04-01



    Type of media :

    Miscellaneous


    Type of material :

    No indication


    Language :

    English





    Conditional Probability Estimation Using Random Kernels

    Tsukamoto, Taro | British Library Online Contents | 2015


    Object Indexing Using an Iconic Sparse Distributed Memory

    Rao, R. / Ballard, D. / IEEE Computer Society et al. | British Library Conference Proceedings | 1995


    Object indexing using an iconic sparse distributed memory

    Rao, R.P.N. / Ballard, D.H. | IEEE | 1995


    Approximation of Failure Probability Using Conditional Sampling

    Giesy, Daniel / Crespo, Luis / Kenny, Sean | AIAA | 2008