These lectures, presented at the 2022 Les Houches Summer School on
Stati...
Although machine learning models typically experience a drop in performa...
The test loss of well-trained neural networks often follows precise powe...
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...
The choice of initial learning rate can have a profound effect on the
pe...
A longstanding goal in deep learning research has been to precisely
char...
There is a previously identified equivalence between wide fully connecte...
In recent years, state-of-the-art methods in computer vision have utiliz...
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...