Autoassociative Memory with high Storage Capacity
نویسندگان
چکیده
The general neural unit (GNU) 1] is known for its high storage capacity as an autoassociative memory. The exponential increase in its storage capacity with the number of inputs per neuron is far greater than the linear growth in the famous Hoppeld network 2]. This paper shows that the GNU attains an even higher capacity with the use of pyramids of neurons instead of single neurons as its nodes. The paper also shows that the storage capacity/cost ratio increases, giving further support to this node upgrade. This analysis combines the modular approach for storage capacity assessment of pyramids 3] and of GNUs 4].
منابع مشابه
Brouwerian Autoassociative Morphological Memories and Their Relation to Traditional and Sparsely Connected Autoassociative Morphological Memories
Autoassociative morphological memories (AMM) are memory models that use operations of mathematical morphology for the storage and recall of pattern associations. These models can be very well de ned in a mathematical structure called complete lattice. In this paper, we introduce the Brouwerian autoassociative morphological memories (BAMMs) that are de ned on a complete Brouwerian lattice. The a...
متن کاملA Capacity Study of a Bayesian Neural Network with Hypercolumns
The dynamics and storage capacity of a Bayesian Confidence Propagation Neural Network (BCPNN) with hypercolumns is studied. Hypercolumns is a concept derived from the local connectivity seen between neurons in cortex. The BCPNN has recurrent connections and it is used as an autoassociative memory. A parallel implementation of the BCPNN is used to run simulations of large BCPNN. Those simulation...
متن کاملThe storage capacity of Potts models for semantic memory retrieval
We introduce and analyse a minimal network model of semantic memory in the human brain. The model is a global associative memory structured as a collection of N local modules, each coding a feature, which can take S possible values, with a global sparseness a (the average fraction of features describing a concept). We show that, under optimal conditions, the number cM of modules connected on av...
متن کاملSparsely Connected Autoassociative Fuzzy Implicative Memories for the Reconstruction of Large Gray-Scale Images
Autoassociative fuzzy implicative memories (AFIMs) are models that exhibit optimal absolute storage capacity and an excellent tolerance with respect to incomplete or eroded patterns. Thus, they can be effectively used for the reconstruction of gray-scale images. In practice, however, applications of AFIMs are confined to images of small size due to computational limitations. This paper introduc...
متن کاملAn Effective Associative Memory for Pattern Recognition
Neuron models of associative memory provide a new and prospective technology for reliable date storage and patterns recognition. However, even when the patterns are uncorrelated, the efficiency of most known models of associative memory is low. We developed a new version of associative memory with record characteristics of its storage capacity and noise immunity, which, in addition, is effectiv...
متن کامل