Search Results for author:
Found 4 papers, 0 papers with code
An initial alignment between neural network and target is needed for gradient descent to learn
This paper introduces the notion of ``Initial Alignment'' (INAL) between a neural network at initialization and a target function.
A Johnson--Lindenstrauss Framework for Randomly Initialized CNNs
The celebrated Johnson--Lindenstrauss lemma answers this question for linear fully-connected neural networks (FNNs), stating that the geometry is essentially preserved.
Regularization by Misclassification in ReLU Neural Networks
We study the implicit bias of ReLU neural networks trained by a variant of SGD where at each step, the label is changed with probability $p$ to a random label (label smoothing being a close variant of this procedure).
Bayesian Decision Making in Groups is Hard
We study the computations that Bayesian agents undertake when exchanging opinions over a network.
