Learning by dilution in a neural network
From MaRDI portal
Abstract: A perceptron with N random weights can store of the order of N patterns by removing a fraction of the weights without changing their strengths. The critical storage capacity as a function of the concentration of the remaining bonds for random outputs and for outputs given by a teacher perceptron is calculated. A simple Hebb-like dilution algorithm is presented which in the teacher case reaches the optimal generalization ability.
Recommendations
- A dilution algorithm for neural networks
- scientific article; zbMATH DE number 4201670
- scientific article; zbMATH DE number 1330071
- Learning theory and approximation by neural networks
- scientific article; zbMATH DE number 4211542
- Diffusion learning algorithms for feedforward neural networks
- Neural network training as a dissipative process
- Learning in the multilayer perceptron
- An efficient dilution strategy for constructing sparsely connected neural networks
Cited in
(8)- Stability of the replica symmetric solution in diluted perceptron learning
- scientific article; zbMATH DE number 4204162 (Why is no real title available?)
- A dilution algorithm for neural networks
- Dynamical properties of a randomly diluted neural network with variable activity
- Dilution in Boolean perceptrons that learn from noisy examples
- Diffusion learning algorithms for feedforward neural networks
- An efficient dilution strategy for constructing sparsely connected neural networks
- The statistical physics of learning in under- and over-parameterized layered neural networks
This page was built for publication: Learning by dilution in a neural network
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4255726)
