Open Problem: Optimal Rates for Stochastic Decision-Theoretic Online Learning Under Differentially Privacy

Abstract

For the stochastic variant of decision-theoretic online learning with $K$ actions, $T$ rounds, and minimum gap $\Delta_{\min}$, the optimal, gap-dependent rate of the pseudo-regret is known to be $O \left( \frac{\log K}{\Delta_{\min}} \right)$. We ask to settle the optimal gap-dependent rate for the problem under $\varepsilon$-differential privacy.

Cite

Text

Hu and Mehta. "Open Problem: Optimal Rates for Stochastic Decision-Theoretic Online Learning Under Differentially Privacy." Conference on Learning Theory, 2024.

Markdown

[Hu and Mehta. "Open Problem: Optimal Rates for Stochastic Decision-Theoretic Online Learning Under Differentially Privacy." Conference on Learning Theory, 2024.](https://mlanthology.org/colt/2024/hu2024colt-open/)

BibTeX

@inproceedings{hu2024colt-open,
  title     = {{Open Problem: Optimal Rates for Stochastic Decision-Theoretic Online Learning Under Differentially Privacy}},
  author    = {Hu, Bingshan and Mehta, Nishant A.},
  booktitle = {Conference on Learning Theory},
  year      = {2024},
  pages     = {5330-5334},
  volume    = {247},
  url       = {https://mlanthology.org/colt/2024/hu2024colt-open/}
}