Steps Toward Deep Kernel Methods from Infinite Neural Networks [article]

Tamir Hazan, Tommi Jaakkola
2015 arXiv   pre-print
Contemporary deep neural networks exhibit impressive results on practical problems. These networks generalize well although their inherent capacity may extend significantly beyond the number of training examples. We analyze this behavior in the context of deep, infinite neural networks. We show that deep infinite layers are naturally aligned with Gaussian processes and kernel methods, and devise stochastic kernels that encode the information of these networks. We show that stability results
more » ... y despite the size, offering an explanation for their empirical success.
arXiv:1508.05133v2 fatcat:iwnescjlubenlmntue3uofm3a4