activity
20162022
most citedPractical Gauss-Newton Optimisation for Deep Learning

35 citations · 51 across the 6 of their papers we have counts for

collaborators
Showing stat.MLShow all

5 papers · 1 filter

stat.ML2020

Disentangling by Subspace Diffusion

David Pfau, Irina Higgins, Aleksandar Botev +1

We present a novel nonparametric algorithm for symmetry-based disentangling of data manifolds, the Geometric Manifold Component Estimator (GEOMANCER). GEOMANCER provides a partial…

stat.ML2018

Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting

Hippolyt Ritter, Aleksandar Botev, David Barber

We introduce the Kronecker factored online Laplace approximation for overcoming catastrophic forgetting in neural networks. The method is grounded in a Bayesian online learning fra…

stat.ML201735 cited

Practical Gauss-Newton Optimisation for Deep Learning

Aleksandar Botev, Hippolyt Ritter, David Barber

We present an efficient block-diagonal ap- proximation to the Gauss-Newton matrix for feedforward neural networks. Our result- ing algorithm is competitive against state- of-the-ar…

stat.ML20161 cited

Nesterov's Accelerated Gradient and Momentum as approximations to Regularised Update Descent

Aleksandar Botev, Guy Lever, David Barber

We present a unifying framework for adapting the update direction in gradient-based iterative optimization methods. As natural special cases we re-derive classical momentum and Nes…

stat.ML20162 cited

Dealing with a large number of classes -- Likelihood, Discrimination or Ranking?

David Barber, Aleksandar Botev

We consider training probabilistic classifiers in the case of a large number of classes. The number of classes is assumed too large to perform exact normalisation over all classes.…