Unconstrained Online Learning with Unbounded Losses
Jacobsen, Andrew, Cutkosky, Ashok
–arXiv.org Artificial Intelligence
Algorithms for online learning typically require one or more boundedness assumptions: that the domain is bounded, that the losses are Lipschitz, or both. In this paper, we develop a new setting for online learning with unbounded domains and non-Lipschitz losses. For this setting we provide an algorithm which guarantees $R_{T}(u)\le \tilde O(G\|u\|\sqrt{T}+L\|u\|^{2}\sqrt{T})$ regret on any problem where the subgradients satisfy $\|g_{t}\|\le G+L\|w_{t}\|$, and show that this bound is unimprovable without further assumptions. We leverage this algorithm to develop new saddle-point optimization algorithms that converge in duality gap in unbounded domains, even in the absence of meaningful curvature. Finally, we provide the first algorithm achieving non-trivial dynamic regret in an unbounded domain for non-Lipschitz losses, as well as a matching lower bound. The regret of our dynamic regret algorithm automatically improves to a novel $L^{*}$ bound when the losses are smooth.
arXiv.org Artificial Intelligence
Jul-14-2023
- Country:
- North America
- Canada > Alberta (0.28)
- United States (0.92)
- North America
- Genre:
- Research Report (0.82)
- Industry:
- Education > Educational Setting > Online (0.83)
- Technology: