Infinite width limit has shed light on generalization and optimization
a...
The Neural Tangent Kernel (NTK), defined as Θ_θ^f(x_1, x_2) =
[∂ f(θ, x_...
We introduce repriorisation, a data-dependent reparameterisation which
t...
The effectiveness of machine learning algorithms arises from being able ...
We perform a careful, thorough, and large scale empirical study of the
c...
Recent work has shown that the prior over functions induced by a deep
Ba...
There is a growing amount of literature on the relationship between wide...
There are currently two parameterizations used to derive fixed kernels
c...
Neural Tangents is a library designed to enable research into infinite-w...
There is a previously identified equivalence between wide fully connecte...
In practice it is often found that large over-parameterized neural netwo...
A deep fully-connected neural network with an i.i.d. prior over its
para...
Existing machine translation decoding algorithms generate translations i...
We explore the method of style transfer presented in the article "A Neur...