Scaling laws for the attractors of Hopfield networks
نویسندگان
چکیده
(Reçu le 28 mars 1985, accepte sous forme définitive le 29 mai 1985) Résumé. 2014 Les réseaux d'automates à seuil sont des systèmes dynamiques à structure aléatoire semblables aux verres de spins dont J. Hopfield a proposé l'application comme mémoires associa-tives. Nous établissons les lois d'échelles reliant le nombre maximum d'attracteurs utiles et la distance d'attraction, au nombre des automates du réseau. Notre approche permet aussi un meilleur choix des seuils, ce qui double les performances du réseau en nombre d'attracteurs. Abstract 2014 Networks of threshold automata are random dynamical systems with a large number of attractors, which J. Hopfield proposed to use as associative memories. We establish the scaling laws relating the maximum number of « useful » attractors and the radius of the attraction basin to the number of automata. A by-product of our analysis is a better choice for thresholds which doubles the performances in terms of the maximum number of « useful » attractors.
منابع مشابه
Transient hidden chaotic attractors in a Hopfield neural system
In this letter we unveil the existence of transient hidden coexisting chaotic attractors, in a simplified Hopfield neural network with three neurons. keyword Hopfield neural network; Transient hidden chaotic attractor; Limit cycle
متن کاملOn the Capacity of Hopfield Neural Networks as EDAs for Solving Combinatorial Optimisation Problems
Multi-modal optimisation problems are characterised by the presence of either local sub-optimal points or a number of equally optimal points. These local optima can be considered as point attractors for hill climbing search algorithms. It is desirable to be able to model them either to avoid mistaking a local optimum for a global one or to allow the discovery of multiple equally optimal solutio...
متن کاملNonlinear Dynamics and Chaos in Fractional-Order Hopfield Neural Networks with Delay
A fractional-order two-neuron Hopfield neural network with delay is proposed based on the classic well-known Hopfield neural networks, and further, the complex dynamical behaviors of such a network are investigated. A great variety of interesting dynamical phenomena, including single-periodic,multiple-periodic, and chaoticmotions, are found to exist.The existence of chaotic attractors is verifi...
متن کاملPatent Number : United States Patent
J. J. Hopfield, “Neural Networks and Physical Systems with Emergent Collective Computational Abilities,” Proc. Nat. Acad. Sci., USA, vol. 79, pp. 2254-2258, Apr. 1982. R. J. McEliece, et al., “The Capacity of the Hopfield Associative Memory,” IEEE Transactions on Information Theory, vol. T-33, pp. 461-482, 1987. B. L. Montgomery et al., “Evaluation of the use of Hopfield Neural Network Model as...
متن کاملU . S . Patent Aug .
J. J. Hopfield, “Neural Networks and Physical Systems with Emergent Collective Computational Abilities,” Proc. Nat. Acad. Sci., USA, vol. 79, pp. 2254-2258, Apr. 1982. R. J. McEliece, et al., “The Capacity of the Hopfield Associative Memory,” IEEE Transactions on Information Theory, vol. T-33, pp. 461-482, 1987. B. L. Montgomery et al., “Evaluation of the use of Hopfield Neural Network Model as...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2016