Effect of the initial configuration of weights on the training and function of artificial neural networks

no code implementations • 4 Dec 2020 • R. J. Jesus, M. L. Antunes, R. A. da Costa, S. N. Dorogovtsev, J. F. F. Mendes, R. L. Aguiar

We compare the evolution of the distribution function of this deviation with the evolution of the loss during training.

Weak Multiplex Percolation

no code implementations • 3 Nov 2020 • G. J. Baxter, R. A. da Costa, S. N. Dorogovtsev, J. F. F. Mendes

We describe a generalisation of percolation to multilayer networks: weak multiplex percolation.

Disordered Systems and Neural Networks