Constant Regret, Generalized Mixability, and Mirror Descent
Zakaria Mhammedi, Robert C. Williamson
–Neural Information Processing Systems
Under this setting, and for the right choice of loss function and "mixing" algorithm, it is possible for the learner to achieve a constant regret regardless of the number of prediction rounds.
Neural Information Processing Systems
Nov-18-2025, 12:21:54 GMT
- Country:
- North America
- Canada > Quebec
- Montreal (0.04)
- United States > New Jersey
- Mercer County > Princeton (0.04)
- Canada > Quebec
- North America
- Technology: