
Curvature is Key: SubSampled Loss Surfaces and the Implications for Large Batch Training
We study the effect of minibatching on the loss landscape of deep neura...
read it

Flatness is a False Friend
Hessian based measures of flatness, such as the trace, Frobenius and spe...
read it

Beyond Random Matrix Theory for Deep Networks
We investigate whether the Wigner semicircle and MarcenkoPastur distri...
read it

Iterate Averaging Helps: An Alternative Perspective in Deep Learning
Iterate averaging has a rich history in optimisation, but has only very ...
read it

MLRG Deep Curvature
We present MLRG Deep Curvature suite, a PyTorchbased, opensource packa...
read it

A Maximum Entropy approach to Massive Graph Spectra
Graph spectral techniques for measuring graph similarity, or for learnin...
read it

MEMe: An Accurate Maximum Entropy Method for Efficient Approximations in LargeScale Machine Learning
Efficient approximation lies at the heart of largescale machine learnin...
read it

Entropic Spectral Learning in Large Scale Networks
We present a novel algorithm for learning the spectral density of large ...
read it

VBALD  Variational Bayesian Approximation of Log Determinants
Evaluating the log determinant of a positive definite matrix is ubiquito...
read it

Entropic Determinants
The ability of many powerful machine learning algorithms to deal with la...
read it

Entropic Trace Estimates for Log Determinants
The scalable calculation of matrix determinants has been a bottleneck to...
read it
Diego Granziol
is this you? claim profile