AISTATS 2024poster1 citations

Understanding Progressive Training Through the Framework of Randomized Coordinate Descent

Rafał Szlendak, Elnur Gasanov, Peter Richtarik

Abstract

We propose a Randomized Progressive Training algorithm (RPT)—a stochastic proxy for the well-known Progressive Training method (PT) (Karras et al., 2017). Originally designed to train GANs (Goodfellow et al., 2014), PT was proposed as a heuristic, with no convergence analysis even for the simplest objective functions. On the contrary, to the best of our knowledge, RPT is the first PT-type algorithm with rigorous and sound theoretical guarantees for general smooth objective functions. We cast our method into the established framework of Randomized Coordinate Descent (RCD) (Nesterov, 2012; Richtarik & Takac, 2014), for which (as a by-product of our investigations) we also propose a novel, simple and general convergence analysis encapsulating strongly-convex, convex and nonconvex objectives. We then use this framework to establish a convergence theory for RPT. Finally, we validate the effectiveness of our method through extensive computational experiments.

BibTeX
@InProceedings{pmlr-v238-szlendak24a,
  title = 	 {Understanding Progressive Training Through the Framework of Randomized Coordinate Descent},
  author =       {Szlendak, Rafa\l{} and Gasanov, Elnur and Richtarik, Peter},
  booktitle = 	 {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {2161--2169},
  year = 	 {2024},
  editor = 	 {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen},
  volume = 	 {238},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {02--04 May},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v238/szlendak24a/szlendak24a.pdf},
  url = 	 {https://proceedings.mlr.press/v238/szlendak24a.html},
  abstract = 	 {We propose a Randomized Progressive Training algorithm (RPT)—a stochastic proxy for the well-known Progressive Training method (PT) (Karras et al., 2017). Originally designed to train GANs (Goodfellow et al., 2014), PT was proposed as a heuristic, with no convergence analysis even for the simplest objective functions. On the contrary, to the best of our knowledge, RPT is the first PT-type algorithm with rigorous and sound theoretical guarantees for general smooth objective functions. We cast our method into the established framework of Randomized Coordinate Descent (RCD) (Nesterov, 2012; Richtarik & Takac, 2014), for which (as a by-product of our investigations) we also propose a novel, simple and general convergence analysis encapsulating strongly-convex, convex and nonconvex objectives. We then use this framework to establish a convergence theory for RPT. Finally, we validate the effectiveness of our method through extensive computational experiments.}
}
Understanding Progressive Training Through the Framework of Randomized Coordinate Descent · AISTATS 2024