arXiv research

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

New framework for Adam-type algorithms with constant β1, improving regret analysis.

Transformers achieve near-optimal dynamic regret in non-stationary reinforcement learning.

Study risk-sensitive reinforcement learning with Lipschitz dynamic risk measures, establishing regret bounds.

Posterior sampling-based EI achieves sublinear regret bounds for expensive function optimization.

This paper analyzes data-driven Newsvendor problems and finds a wide range of possible regrets.

Improved algorithm reduces regret in corrupted expert advice setting.

New algorithm improves bandit with graph feedback by decomposing regret.

We consider the exploration-exploitation tradeoff in linear quadratic (LQ) control problems, where the state dynamics is linear and the cost function is quadratic in states and controls. We analyze the regret of Thompson sampling (TS) (a.k.a. posterior-sampling for reinforcement learning) in the frequentist setting, i.…

Linear contextual bandit is an important class of sequential decision making problems with a wide range of applications to recommender systems, online advertising, healthcare, and many other machine learning related tasks. While there is a lot of prior research, tight regret bounds of linear contextual bandit with infi…

Near-logarithmic regret per switch achieved for mixable/exp-concave losses.

New algorithm reduces regret in noisy context bandits.

New algorithm reduces regret in bandit optimization for high-dimensional data.

First sample-efficient algorithm for learning EFCE in bandit feedback settings.

New algorithm DUSA minimizes regret in structured bandits with structural information.

This paper improves Bayesian optimization methods with tighter regret bounds and practical solutions.

New algorithms minimize regret with global costs in online learning.

Kernel-UCBVI algorithm balances exploration and exploitation in metric state-action spaces.

We study revenue optimization learning algorithms for repeated posted-price auctions where a seller interacts with a single strategic buyer that holds a fixed private valuation for a good and seeks to maximize his cumulative discounted surplus. For this setting, first, we propose a novel algorithm that never decreases …

Supplier learns to price contracts against a learning retailer.

Many prediction domains, such as ad placement, recommendation, trajectory prediction, and document summarization, require predicting a set or list of options. Such lists are often evaluated using submodular reward functions that measure both quality and diversity. We propose a simple, efficient, and provably near-optim…

The problem of multi-armed bandits (MAB) asks to make sequential decisions while balancing between exploitation and exploration, and have been successfully applied to a wide range of practical scenarios. Various algorithms have been designed to achieve a high reward in a long term. However, its short-term performance m…

Paper tackles online learning on curved spaces without projections.

Paper introduces risk-sensitive bandits with optimal arm mixtures.

We study the pricing problem faced by a firm that sells a large number of products, described via a wide range of features, to customers that arrive over time. Customers independently make purchasing decisions according to a general choice model that includes products features and customers' characteristics, encoded as…

Improved ExO method achieves near-optimal bounds in both stochastic and adversarial settings.

Study shows how AI model can improve decision-making with missing data.

Papers learn from data to make decisions without interacting, improving on previous methods.

Predictive sampling improves on Thompson sampling for non-stationary bandit environments.

The paper provides robustness guarantees for mode estimation in bandits.

This work shows neural networks can solve non-convex constraints problems.

Universal algorithm learns unknown distribution for various decision-making problems.

Algorithm learns optimal coordination for strategic agents in uncertain settings.

New algorithm for contextual combinatorial bandits with probabilistic arm triggering.

Sequential tests for nonparametric hypotheses using supermartingales.

Optimal algorithms for mixable losses in dynamic environments with reduced redundancy.

ECPv2 optimizes Lipschitz functions efficiently and scalably.

A new Python-C++ framework for agent-based simulation.

New RL algorithm handles delayed feedback with posterior sampling.

We are interested in a framework of online learning with kernels for low-dimensional but large-scale and potentially adversarial datasets. We study the computational and theoretical performance of online variations of kernel Ridge regression. Despite its simplicity, the algorithm we study is the first to achieve the op…

Bayesian optimization tackles expensive discrete and mixed parameter spaces.

Regret minimization is treated as the golden rule in the traditional study of online learning. However, regret minimization algorithms tend to converge to the static optimum, thus being suboptimal for changing environments. To address this limitation, new performance measures, including dynamic regret and adaptive regr…

The paper analyzes the sliding regret of stochastic bandit algorithms.

This paper analyzes regret bounds for Gaussian process Thompson sampling.

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 …

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…

Balances the regret of different algorithms in bandit and RL problems.

Data-driven model selection reduces regret in sequential decisions.