Now showing items 1-2 of 2

  • Orvieto, Antonio; Kersting, Hans; Proske, Frank Norbert; Bach, Francis; Lucchi, Aurelien (Journal article / Tidsskriftartikkel / PublishedVersion; Peer reviewed, 2022)
    Injecting artificial noise into gradient descent (GD) is commonly employed to improve the performance of machine learning models. Usually, uncorrelated noise is used in such perturbed gradient descent (PGD) methods. It is, ...
  • Lucchi, Aurelien; Proske, Frank Norbert; Orvieto, Antonio; Bach, Francis; Kersting, Hans (Journal article / Tidsskriftartikkel / AcceptedVersion; Peer reviewed, 2022)
    Studying the properties of stochastic noise to optimize complex non-convex functions has been an active area of research in the field of machine learning. Prior work~\citep{zhou2019pgd, wei2019noise} has shown that the ...