Better Full-Matrix Regret via Parameter-Free Online Learning

Abstract

We provide online convex optimization algorithms that guarantee improved full-matrix regret bounds. These algorithms extend prior work in several ways. First, we seamlessly allow for the incorporation of constraints without requiring unknown oracle-tuning for any learning rate parameters. Second, we improve the regret of the full-matrix AdaGrad algorithm by suggesting a better learning rate value and showing how to tune the learning rate to this value on-the-fly. Third, all our bounds are obtained via a general framework for constructing regret bounds that depend on an arbitrary sequence of norms.

Cite

Text

Cutkosky. "Better Full-Matrix Regret via Parameter-Free Online Learning." Neural Information Processing Systems, 2020.

Markdown

[Cutkosky. "Better Full-Matrix Regret via Parameter-Free Online Learning." Neural Information Processing Systems, 2020.](https://mlanthology.org/neurips/2020/cutkosky2020neurips-better/)

BibTeX

@inproceedings{cutkosky2020neurips-better,
  title     = {{Better Full-Matrix Regret via Parameter-Free Online Learning}},
  author    = {Cutkosky, Ashok},
  booktitle = {Neural Information Processing Systems},
  year      = {2020},
  url       = {https://mlanthology.org/neurips/2020/cutkosky2020neurips-better/}
}