On a Frank-Wolfe Approach for Abs-smooth Optimization ongoing

Motivated by nonsmooth problems in machine learning, we solve the problem of minimizing an abs-smooth function subject to closed convex constraints. New theory and algorithms are developed using linear minimization oracles to enforce constraints and abs-linearization methods to handle nonsmoothness.

🧑‍🎓 Project Members

Andrea Walther
Principal Investigator
Sebastian Pokutta
Principal Investigator
pokutta (at) zib.de
Zev Woodstock
Principal Investigator
woodstock (at) zib.de

🪙 Funding

This project is being funded by the Berlin Mathematics Research Center MATH+ (project ID EF1-23), itself funded by the German Research Foundation (DFG) under Germany's Excellence Strategy (EXC-2046/1, project ID 390685689) from April 2023 to March 2026.

<📝 Publications and preprints

  1. Kreimeier, T., Pokutta, S., Walther, A., and Woodstock, Z. (2023). On a Frank-Wolfe Approach for Abs-smooth Functions. Optimization Methods and Software. [arXiv]
    [BibTeX]
    @article{2023_KreimeierPokuttaWaltherWoodstock_Frankwolfeabssmooth,
      year = {2023},
      journal = {Optimization Methods and Software},
      archiveprefix = {arXiv},
      eprint = {2303.09881},
      primaryclass = {math.OC},
      author = {Kreimeier, Timo and Pokutta, Sebastian and Walther, Andrea and Woodstock, Zev},
      title = {On a Frank-Wolfe Approach for Abs-smooth Functions}
    }
  2. Woodstock, Z., and Pokutta, S. (2023). Splitting the Conditional Gradient Algorithm. [arXiv]
    [BibTeX]
    @misc{2023_WoodstockPokutta_Conditionalgradientnonconvex,
      archiveprefix = {arXiv},
      eprint = {2311.05381},
      primaryclass = {math.OC},
      year = {2023},
      author = {Woodstock, Zev and Pokutta, Sebastian},
      title = {Splitting the Conditional Gradient Algorithm}
    }