Research data supporting the publication "Choose your path wisely: gradient descent in a Bregman distance framework"
Repository URI
Repository DOI
Change log
Authors
Betcke, Marta
Description
This is the research data that accompanies the publication "Choose your path wisely: gradient descent in a Bregman distance framework". It contains code for all numerical examples given in the numerical results section of the associated publication. Note that the parallel MRI example and the classification example require additional external data. The corresponding URLs can be found in the associated publication
Version
Software / Usage instructions
This software requires MATLABR2016b or higher. Once the files are unzipped into a folder, run the file setpath.m in order to add the relevant folders to the MATLAB path. All numerical experiments are contained in the folder 'Examples'.
Keywords
Nonconvex optimisation, Nonsmooth optimisation, Gradient descent, Bregman iteration, Linearised Bregman iteration, Parallel MRI, Blind deconvolution, Deep learning
Publisher
Sponsorship
Engineering and Physical Sciences Research Council (EP/M00483X/1)
Leverhulme Trust (ECF-2016-611)
Isaac Newton Trust (1608(aj))
Engineering and Physical Sciences Research Council (EP/N014588/1)
Engineering and Physical Sciences Research Council (EP/J009539/1)
Alan Turing Institute (unknown)
European Commission Horizon 2020 (H2020) Marie Sk?odowska-Curie actions (691070)
Leverhulme Trust (ECF-2016-611)
Isaac Newton Trust (1608(aj))
Engineering and Physical Sciences Research Council (EP/N014588/1)
Engineering and Physical Sciences Research Council (EP/J009539/1)
Alan Turing Institute (unknown)
European Commission Horizon 2020 (H2020) Marie Sk?odowska-Curie actions (691070)
This work was funded by the Leverhulme Trust Early Career Fellowship ’Learning from mistakes: a supervised feedback-loop for imaging applications’, the Isaac Newton Trust, the Engineering and Physical Sciences Research Council (EPSRC) ’EP/K009745/1’, the Leverhulme Trust project ’Breaking the non-convexity barrier’, the EPSRC grant ’EP/M00483X/1’, the EPSRC centre ’EP/N014588/1’, the Cantab Capital Institute for the Mathematics of Information and CHiPS (Horizon 2020 RISE project grant).