2 problems
- 0 votes0 replies2 views
Extension of the expected-regret limit to general sample spaces
Let be a sample space, and let , , and be as in the paper's main theorem, with the model parameter associated with the distribution in the…
- 0 votes0 replies0 views
Extension of the server-idleness regret lower bound to nonintegral decisions
Nonintegral-decision extension conjecture. A similar regret lower bound should hold even without the integrality assumption on the online algorithm's decisions.