### Refine

#### Document Type

- Article (2)
- In Proceedings (2)

#### Language

- English (4)

#### Has Fulltext

- no (4)

#### Is part of the Bibliography

- no (4)

#### Institute

Several learning problems involve solving min-max problems, e.g., empirical distributional robust learning
[Namkoong and Duchi, 2016, Curi et al., 2020] or learning with non-standard aggregated losses [Shalev-
Shwartz and Wexler, 2016, Fan et al., 2017]. More specifically, these problems are convex-linear problems
where the minimization is carried out over the model parameters w ∈ W and the maximization over the
empirical distribution p ∈ K of the training set indexes, where K is the simplex or a subset of it. To design
efficient methods, we let an online learning algorithm play against a (combinatorial) bandit algorithm.
We argue that the efficiency of such approaches critically depends on the structure of K and propose two
properties of K that facilitate designing efficient algorithms. We focus on a specific family of sets Sn,k
encompassing various learning applications and provide high-probability convergence guarantees to the
minimax values.

In this work, we analyze two of the most fundamental algorithms in geodesically convex optimization: Riemannian gradient descent and (possibly inexact) Riemannian proximal point. We quantify their rates of convergence and produce different variants with several trade-offs. Crucially, we show the iterates naturally stay in a ball around an optimizer, of radius depending on the initial distance and, in some cases, on the curvature. Previous works simply assumed bounded iterates, resulting in rates that were not fully quantified. We also provide an implementable inexact proximal point algorithm and prove several new useful properties of Riemannian proximal methods: they work when positive curvature is present, the proximal operator does not move points away from any optimizer, and we quantify the smoothness of its induced Moreau envelope. Further, we explore beyond our theory with empirical tests.

Linear bandit algorithms yield O~(n√T) pseudo-regret bounds on compact convex action sets K⊂Rn and two types of structural assumptions lead to better pseudo-regret bounds. When K is the simplex or an ℓp ball with p∈]1,2], there exist bandits algorithms with O~(√n√T) pseudo-regret bounds. Here, we derive bandit algorithms for some strongly convex sets beyond ℓp balls that enjoy pseudo-regret bounds of O~(√n√T), which answers an open question from [BCB12, §5.5.]. Interestingly, when the action set is uniformly convex but not necessarily strongly convex, we obtain pseudo-regret bounds with a dimension dependency smaller than O(√n). However, this comes at the expense of asymptotic rates in T varying between O(√T) and O(T).