Search Results for author:
Found 1 papers, 0 papers with code
Wide neural networks: From non-gaussian random fields at initialization to the NTK geometry of training
In particular, using a certain metric in the space of measures we find that, along training, the resulting measure is within $n^{-\frac{1}{2}}(\log n)^{1+}$ of the time dependent Gaussian process corresponding to the infinite width network (which is explicitly given by precomposing the initial Gaussian process with the linear flow corresponding to training in the infinite width limit).
Cannot find the paper you are looking for? You can
Submit a new open access paper.
