arXiv research

Efficient algorithms for online learning with changing action sets, achieving no-approximate-regret guarantees.

Paper solves no-swap regret minimization for combinatorial bandits with polylogarithmic dependence on N.

No-regret learning with strategic experts, incentivized.

No-regret optimization for time-varying functions using uncertainty injection.

The notion of \emph{policy regret} in online learning is a well defined? performance measure for the common scenario of adaptive adversaries, which more traditional quantities such as external regret do not take into account. We revisit the notion of policy regret and first show that there are online learning settings …

New insights link no-regret learning to online conformal prediction in adversarial settings.

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

Paper proposes OPF policy for fair resource allocation with sublinear regret.

We introduce CSE for MLSF games and devise online learning algorithms for achieving no-external Stackelberg-regret.

This paper tackles no-regret learning for fair multi-agent social welfare optimization.

Calibrated strategies can be obtained by performing strategies that have no internal regret in some auxiliary game. Such strategies can be constructed explicitly with the use of Blackwell's approachability theorem, in an other auxiliary game. We establish the converse: a strategy that approaches a convex $B$-set can be…

No-regret learning fails to converge to Nash equilibria in mixed strategies.

Paper analyzes GP-EI for Bayesian optimization with no regret and provides guidance on choosing incumbents.

The paper analyzes the sliding regret of stochastic bandit algorithms.

Paper establishes no-regret property for practical EGO optimization.

New algorithms achieve no-regret learning even with adversarial transitions and losses.

New framework uses PDE for no-regret generative modeling.

No communication allows optimal instance-dependent regret guarantees in multi-player bandits.

New research shows no-regret learning is impossible in Markov games under certain assumptions.

Paper proposes no-regret algorithms for private GP bandit optimization.

The paper tackles robust policy learning from multiple data sources.

New strategy achieves optimal regret without communication or collisions in multi-player bandit.

There are two variants of the classical multi-armed bandit (MAB) problem that have received considerable attention from machine learning researchers in recent years: contextual bandits and simple regret minimization. Contextual bandits are a sub-class of MABs where, at every time step, the learner has access to side in…

A new mechanism reduces expert belief regret in online forecasting.

We consider the classical stochastic multi-armed bandit but where, from time to time and roughly with frequency $ε$, an extra observation is gathered by the agent for free. We prove that, no matter how small $ε$ is the agent can ensure a regret uniformly bounded in time. More precisely, we construct an algorithm with a…

Counterfactual Regret Minimization (CFR) has found success in settings like poker which have both terminal states and perfect recall. We seek to understand how to relax these requirements. As a first step, we introduce a simple algorithm, local no-regret learning (LONR), which uses a Q-learning-like update rule to allo…

Framework for games with uncertain parameters, ensuring no player can improve by changing strategy.

Kernel-based function approximation improves reinforcement learning performance.

Optimistic Thompson Sampling reduces regret in unknown multi-player games.

LIBO optimizes repeated bandit tasks without prior knowledge or regret.

New algorithm for multi-player bandits with collision-dependent rewards.

Algorithm reduces regret in safe Bayesian optimization with monotonicity constraints.

We consider a family of learning strategies for online optimization problems that evolve in continuous time and we show that they lead to no regret. From a more traditional, discrete-time viewpoint, this continuous-time approach allows us to derive the no-regret properties of a large class of discrete-time algorithms i…

Memory-limited learning tackles adversarial bandits with reduced storage.

Sequential prediction problems such as imitation learning, where future observations depend on previous predictions (actions), violate the common i.i.d. assumptions made in statistical learning. This leads to poor performance in theory and often in practice. Some recent approaches provide stronger guarantees in this se…

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 method tackles high-dimensional contextual bandits with flexible kernel models.

DORIS algorithm achieves no-regret learning in Markov games with adversarial opponents.

Online learning algorithms are designed to learn even when their input is generated by an adversary. The widely-accepted formal definition of an online algorithm's ability to learn is the game-theoretic notion of regret. We argue that the standard definition of regret becomes inadequate if the adversary is allowed to a…

The paper tackles cooperative RL with function approximation, achieving near-optimal learning with limited communication.

New algorithm tackles smooth online learning with optimal regret.

Mirror descent with an entropic regularizer is known to achieve shifting regret bounds that are logarithmic in the dimension. This is done using either a carefully designed projection or by a weight sharing technique. Via a novel unified analysis, we show that these two approaches deliver essentially equivalent bounds …

This paper proposes a new portfolio allocation method using LLMs to outperform traditional strategies.

Algorithms for hyperparameter optimization abound, all of which work well under different and often unverifiable assumptions. Motivated by the general challenge of sequentially choosing which algorithm to use, we study the more specific task of choosing among distributions to use for random hyperparameter optimization.…

We prove a new minimax theorem connecting the worst-case Bayesian regret and minimax regret under partial monitoring with no assumptions on the space of signals or decisions of the adversary. We then generalise the information-theoretic tools of Russo and Van Roy (2016) for proving Bayesian regret bounds and combine th…

Regret analysis is challenging in Multi-Agent Reinforcement Learning (MARL) primarily due to the dynamical environments and the decentralized information among agents. We attempt to solve this challenge in the context of decentralized learning in multi-agent linear-quadratic (LQ) dynamical systems. We begin with a simp…

New algorithm for learning preferences in decentralized matching markets reduces regret to logarithmic levels.

Algorithm learns expert weights to minimize regret in adversarial setting.