arXiv research

Paper explores rate-preserving reductions between Blackwell approachability and no-regret learning.

Paper analyzes faster convergence rates for reinforcement learning from offline data.

New adaptive learning rate for FTRL reduces regret to Θ(T^2/3).

The paper tightens the regret rate for linear bandit problems.

Optimal learning rate schedules for SGD in changing data distributions.

New algorithm optimizes best arm identification with minimal regret.

New bounds show polyhedral surrogates are optimal for generalization.

New algorithm optimizes beam and rate allocation in mmWave systems for multiple users.

Recent work on follow the perturbed leader (FTPL) algorithms for the adversarial multi-armed bandit problem has highlighted the role of the hazard rate of the distribution generating the perturbations. Assuming that the hazard rate is bounded, it is possible to provide regret analyses for a variety of FTPL algorithms f…

The paper improves evolutionary computation by optimizing selection rates.

Optimistic Hedge achieves optimal regret bounds in two-player zero-sum games.

We consider the use of no-regret algorithms to compute equilibria for particular classes of convex-concave games. While standard regret bounds would lead to convergence rates on the order of $O(T^{-1/2})$, recent work \citep{RS13,SALS15} has established $O(1/T)$ rates by taking advantage of a particular class of optimi…

New algorithm reduces best-in-class regret in contextual bandits.

Kernel $ε$-Greedy optimizes multi-armed bandits with covariates for sub-linear regret.

Adaptive smooth non-stationary bandits achieve optimal regret rates without knowing parameters.

Policy gradient algorithm with variable learning rates achieves near-optimal performance in multi-arm bandit problems.

New method improves CATE model selection with optimal regret rates.

We study a nonparametric contextual bandit problem where the expected reward functions belong to a Hölder class with smoothness parameter $β$. We show how this interpolates between two extremes that were previously studied in isolation: non-differentiable bandits ($β\leq1$), where rate-optimal regret is achieved by run…

We consider the classical stochastic multi-armed bandit problem with a constraint that limits the total cost incurred by switching between actions to be no larger than a given switching budget. For this problem, we prove matching upper and lower bounds on the optimal (i.e., minimax) regret, and provide efficient rate-o…

Study on online nonparametric regression using Sobolev kernel methods.

New reinforcement learning algorithm achieves instance-optimal sample complexity.

We study the problem of online learning with non-convex losses, where the learner has access to an offline optimization oracle. We show that the classical Follow the Perturbed Leader (FTPL) algorithm achieves optimal regret rate of $O(T^{-1/2})$ in this setting. This improves upon the previous best-known regret rate of…

Bayesian optimisation algorithm for unknown search spaces with sub-linear regret.

Logarithmic regret for continuous-time reinforcement learning.

Study on policy gradient for stochastic bandits using diffusion approximation.

Decentralized learning ensures stability in online queuing systems with packet rates above 1.

Paper characterizes minimax regret rates for online ranking with top-k feedback.

Paper eliminates warm-up phase for PO in linear MDPs, achieving optimal regret.

Study minimax regret in bilateral trade with heavy-tailed valuations.

The paper tackles adaptive policy selection to maximize social welfare, achieving optimal regret bounds.

This guide simplifies high-probability regret bounds in empirical risk minimization.

The study analyzes how covariance estimation errors affect the global minimum-variance portfolio under heavy-tailed distributions.

RandQL is a new model-free algorithm for MDPs with a novel learning rate randomization approach.

Bayesian bandit algorithms with approximate inference improve regret bounds in stochastic linear bandits.

A new metric, Weighted Regret, unifies FDR and power evaluation in online multiple testing.

The study compares uniform-price and discriminatory auctions in terms of learning difficulty.

IDS optimizes regret in stochastic partial monitoring with linear rewards.

This paper establishes that optimistic algorithms attain gap-dependent and non-asymptotic logarithmic regret for episodic MDPs. In contrast to prior work, our bounds do not suffer a dependence on diameter-like quantities or ergodicity, and smoothly interpolate between the gap dependent logarithmic-regret, and the $\wid…

Online prediction improves with abstention, reducing regret to a fixed bound.

Study on optimal rates for sequential probability assignment using smoothed analysis.

Improved cumulative regret for sequence prediction with limited expert advice.

Optimal switching regret for all segmentations in online convex optimisation.

In this paper, we formalise order-robust optimisation as an instance of online learning minimising simple regret, and propose Vroom, a zero'th order optimisation algorithm capable of achieving vanishing regret in non-stationary environments, while recovering favorable rates under stochastic reward-generating processes.…

OMD and DA perform similarly in static settings but OMD is inferior under dynamic learning rates.

New algorithms minimize dynamic regret in non-stationary online learning.

New algorithm reduces adaptation lag in online model selection.

Algorithm reduces regret in distributed kernel bandits with shared randomness.

We study the problem of online learning with a notion of regret defined with respect to a set of strategies. We develop tools for analyzing the minimax rates and for deriving regret-minimization algorithms in this scenario. While the standard methods for minimizing the usual notion of regret fail, through our analysis …