Thermodynamic properties of the<i>Q</i>-state Potts-glass neural network
Physical Review APublished 1 March 1992
D. Bollé, Patrick Dupont, J. Huyghebaert
Citations39
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The Q-state Potts model of neural networks, extended to include biased patterns, is studied for extensive loading and a tricritical point is found in the spin-glass transition for Qg6, depending on \ensuremath{\alpha}.
Abstract
The Q-state Potts model of neural networks, extended to include biased patterns, is studied for extensive loading \ensuremath{\alpha}. Within the replica-symmetric approximation, mean-field equations are written down for general Q and arbitrary temperature T. The critical storage capacity is discussed for Q=3 and two classes of representative bias parameters. The complete T-\ensuremath{\alpha} phase diagram is presented. A tricritical point is found in the spin-glass transition for Q>6, depending on \ensuremath{\alpha}. Contrary to the Hopfield model, the critical lines do not converge to the same T as \ensuremath{\alpha}\ensuremath{\rightarrow}0. A stability analysis is made.
Keywords
Computer SciencePhysics and Astronomy
Physical Review LettersSolvable Model of a Spin-Glass
4,298 Citations1975David C. Sherrington, Scott Kirkpatrick
Journal of Physics A Mathematical and GeneralStability of the Sherrington-Kirkpatrick solution of a spin glass model
2,043 Citations1978J. R. L. de Almeida, D. J. Thouless
Physical Review LettersStoring Infinite Numbers of Patterns in a Spin-Glass Model of Neural Networks
1,047 Citations1985Daniel J. Amit, Hanoch Gutfreund +1 more
The Hopfield model for a neural network is studied in the limit when the number of stored patterns increases with the size of the network, as $p=\ensuremath{\alpha}N$, and it is shown that, despite its spin-glass features, the model exhibits associative memory.
Physical review. B, Condensed matterInfinite-ranged models of spin-glasses
981 Citations1978Scott Kirkpatrick, David C. Sherrington
Annals of PhysicsStatistical mechanics of neural networks near saturation
849 Citations1987Daniel J. Amit, Hanoch Gutfreund +1 more
The Hopfield model of a neural network near its saturation is studied, and it is found that a field coupled to many patterns has a very limited utility in enhancing their learning.
Physical Review LettersMean-field theory of the Potts glass
368 Citations1985David J. Gross, Ido Kanter +1 more
Randomly interacting p-state Potts spins may freeze into a Potts-glass phase in which the Potts symmetry is unbroken, on the average, and the mean-field theory of this phase transition is presented.
Physical review. A, General physicsInformation storage in neural networks with low levels of activity
344 Citations1987Daniel J. Amit, Hanoch Gutfreund +1 more
The Hopfield model of a neural network is extended to allow for the storage and retrieval of biased patterns, and a global constraint is implemented, which restricts the configuration space to states whose mean activity is in the neighborhood of (1/2)(1+a).
Physical review. A, General physicsDiscrete-state phasor neural networks
105 Citations1988André J. Noest
An associative memory network with local variables assuming one of £q$ equidistant positions on the unit circle ($q$-state phasors) is introduced, and its recall behavior is solved exactly for any $q$ when the interactions are sparse and asymmetric.
Journal of Physics A Mathematical and GeneralThe mean-field theory of a Q-state neural network model
103 Citations1989J. Cook
The mean-field equations of a Q-state clock neural network model are derived in the replica-symmetric approximation using the replica method and it is shown that an infinite number of patterns can be stored in the network even in the limit Q to infinity.
International Journal of Modern Physics BTHE AUGMENTED MODELS OF ASSOCIATIVE MEMORY ASYMMETRIC INTERACTION AND HIERARCHY OF PATTERNS
72 Citations1987M. V. Feigel’man, L. B. Ioffe
It is shown that the asymmetry of the Hopfield model of the associative memory does not change the main properties of the model, but leads to the internal nonthermal noise.
Journal of Physics A Mathematical and GeneralStoring an extensive number of grey-toned patterns in a neural network using multistate neurons
56 Citations1990Heiko Rieger
It is shown that it can be done with networks that use neurons which have only a discrete set of possible activities to stabilize grey tones, and the pseudo-inverse rule for the synaptic couplings is demonstrated.
Journal of Physics A Mathematical and GeneralNeural networks that use three-state neurons
44 Citations1989Jonathan S. Yedidia
The dynamics of a neural network which uses three-state neurons (1, 0 and -1) is solved exactly in the limit of non-symmetric and highly dilute synapses and mixture states which have an overlap with two patterns can be stabilised by a threshold in the networks using three- state neurons.
Journal of Physics A Mathematical and GeneralStability properties of Potts neural networks with biased patterns and low loading
41 Citations1991D. Bollé, Patrick Dupont +1 more
Journal of Physics A Mathematical and GeneralLearning grey-toned patterns in neural networks
34 Citations1991Stephan Mertens, Heinrich Köhler +1 more
A fast converging, adaptive learning algorithm is presented, which finds the coupling matrix of optimal stability, which yields the distribution of local fields, the critical storage capacity alpha c and the minimum number of errors for an overloaded network.
Zeitschrift für Physik B Condensed MatterMartingale approach to neural networks with hierarchically structured information
25 Citations1988S. Bös, Reimer Kühn +1 more
A modified version of a model with hierarchically structured information, originally proposed by Parga and Virasoro, is studied in the context of martingale theory and the free energy greatly simplifies and for the retrieval states is shown to reduce to the one of the Hopfield model.
Journal de Physique IStorage capacity of a Potts-perceptron
21 Citations1991Jean‐Pierre Nadal, Albrecht Rau
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not, for teaching and research institutions in France or abroad, or from public or private research centers.
Europhysics Letters (EPL)The Optimal Storage Capacity for a Neural Network with Multi-State Neurons
17 Citations1991D. Bollé, Patrick Dupont +1 more
Using the Gardner approach, the optimal storage capacity is discussed for a network with multi-stage neurons and a spherical constraint on the couplings and it is found that this capacity decreases with the number of states and the corresponding separation of the plateaus of the gain function.
Journal of Physics A Mathematical and GeneralImage evolution in Potts-glass neural networks
17 Citations1989D. Bollé, F. Mallezie
Neural networks of the Potts type, where neurons can occupy q>or=2 discrete states, are considered for couplings which need not be symmetric.
Physical Review APotts-glass model of layered feedforward neural networks
12 Citations1992G. M. Shim, Man-Sun Kim +1 more