papers AI Learner
The Github is limit! Click to go to the new site.

A Neural Network Assembly Memory Model Based on an Optimal Binary Signal Detection Theory

2003-09-21
Petro M. Gopych

Abstract

A ternary/binary data coding algorithm and conditions under which Hopfield networks implement optimal convolutional or Hamming decoding algorithms has been described. Using the coding/decoding approach (an optimal Binary Signal Detection Theory, BSDT) introduced a Neural Network Assembly Memory Model (NNAMM) is built. The model provides optimal (the best) basic memory performance and demands the use of a new memory unit architecture with two-layer Hopfield network, N-channel time gate, auxiliary reference memory, and two nested feedback loops. NNAMM explicitly describes the dependence on time of a memory trace retrieval, gives a possibility of metamemory simulation, generalized knowledge representation, and distinct description of conscious and unconscious mental processes. A model of smallest inseparable part or an “atom” of consciousness is also defined. The NNAMM’s neurobiological backgrounds and its applications to solving some interdisciplinary problems are shortly discussed. BSDT could implement the “best neural code” used in nervous tissues of animals and humans.

Abstract (translated by Google)
URL

https://arxiv.org/abs/cs/0309036

PDF

https://arxiv.org/pdf/cs/0309036


Similar Posts

Comments