E. Gardner and B. Derrida. Three unfinished works on the optimal storage capacity of networks. J. Phys. A: Math. Gen., 22:1983. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Storage Capacity of Letter Recognition in Hopfield Networks - Gang Wei Gwei (Correct)
....have been proposed to escape the local minima and increase the storage capacity. Kohonen et al. 10] proposed a new learning mechanism for non zero self connection networks called pseudo inverse learning rule by extending the Hebbian learning rule, and enlarged the capacity to p = N. Gardner [11] showed that the ultimate capacity will be p = 2N as the basin size tend to zero. 4 EXPERIMENT All experiments in this paper are carried out on the model of letter patterns. The 26 letter patterns are presented with binary representation (0 or 1) of 156 neurons. The synaptic weight values are ....
E. Gardner and B. Derrida. Three unfinished works on the optimal storage capacity of networks. 3. Phys. A: Math. Gen., 22:1983.
Generativity and Systematicity in Neural Network Combinatorial.. - Brousse (1993) (8 citations) (Correct)
....to the problem that have originated from the field of statistical mechanics. When viewed as collections of simple interacting units, neural networks closely resemble large scale atomic physical systems, and their dynamics can, indeed, be studied in a similar way (Tishby et al. 1989) Drawing on (Gardner and Derrida, 1989) who used mean field analysis to study optimal storage capacity of Hopfield networks, a calculation of the generalization ability as a function of the size of the randomly chosen training set, this independently of the learning algorithm used, was possible (Opper et al. 1990) In (Krogh and ....
E. Gardner and B. Derrida. Three unfinished works on the optimal storage capacity of networks. Journal of physics A, 22:1983--1994, 1989.
Rigorous Learning Curve Bounds from Statistical Mechanics - Haussler, Kearns, Seung.. (1996) (40 citations) (Correct)
....near 0, and this is discussed in Section 3.5. 2.6. Analysis of the Ising perceptron We now tackle some real examples of the application of our theory, complete with determination of the appropriate scaling function and a permissible entropy bound. We first consider the class of Ising perceptrons (Gardner Derrida, 1989; Gyorgyi, 1990; Sompolinsky et al. 1990) Suppose that the function class F N consists of all homogeneous perceptrons in which the weights are constrained to be 1 5 . Let the distribution D N be any spherically symmetric distribution on # N , and let the target function f N # F N be ....
....s(#) # log(1 #) for the Ising perceptron, plotted for the same values of # 1 , # 2 , # 3 as in figure 3. reached zero error with probability approaching 1 in the thermodynamic limit for scaled sample size greater than 1.448. This bound on the critical value was known from the work of Gardner and Derrida (1989), and extended to the case of boolean inputs by Baum, Lyuu and Rivin (1991; 1992) Here we are actually giving a bound on the entire learning curve, and the behavior of our bound is very similar in shape to learning curves obtained in both simulations and non rigorous replica calculations from ....
Gardner, E., & Derrida, B. (1989). Three unfinished works on the optimal storage capacity of networks.
Efficient Noise-Tolerant Learning From Statistical Queries - Kearns (1998) (100 citations) (Correct)
No context found.
E. Gardner and B. Derrida. Three unfinished works on the optimal storage capacity of networks. J. Phys. A: Math. Gen., 22:1983.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC