Search Results for author:
Found 4 papers, 0 papers with code
U-Clip: On-Average Unbiased Stochastic Gradient Clipping
U-Clip is a simple amendment to gradient clipping that can be applied to any iterative gradient optimization algorithm.
Provably Strict Generalisation Benefit for Invariance in Kernel Methods
It is a commonly held belief that enforcing invariance improves generalisation.
Provably Strict Generalisation Benefit for Equivariant Models
It is widely believed that engineering a model to be invariant/equivariant improves generalisation.
Lottery Tickets in Linear Models: An Analysis of Iterative Magnitude Pruning
We analyse the pruning procedure behind the lottery ticket hypothesis arXiv:1803. 03635v5, iterative magnitude pruning (IMP), when applied to linear models trained by gradient flow.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
