Recent advances in generative artificial intelligence (GenAI) models have enabled the generation of personalized content that adapts to up-to-date user context. While personalized decision systems are often modeled using bandit formulations, the integration of GenAI introduces new structure into otherwise classical sequential learning problems. In GenAI-powered interventions, the agent selects a query, but the environment experiences a stochastic response drawn from the generative model. Standard bandit methods do not explicitly account for this structure, where actions influence rewards only through stochastic, observed treatments. We introduce generator-mediated bandit-Thompson sampling (GAMBITTS), a bandit approach designed for this action/treatment split, using mobile health interventions with large language model-generated text as a motivating case study. GAMBITTS explicitly models both the treatment and reward generation processes, using information in the delivered treatment to accelerate policy learning relative to standard methods. We establish regret bounds for GAMBITTS by decomposing sources of uncertainty in treatment and reward, identifying conditions where it achieves stronger guarantees than standard bandit approaches. In simulation studies, GAMBITTS consistently outperforms conventional algorithms by leveraging observed treatments to more accurately estimate expected rewards.

We propose a new adaptive sampling approach to multiple testing which aims to maximize statistical power while ensuring anytime false discovery control. We consider $n$ distributions whose means are partitioned by whether they are below or equal to a baseline (nulls), versus above the baseline (true positives). In addition, each distribution can be sequentially and repeatedly sampled. Using techniques from multi-armed bandits, we provide an algorithm that takes as few samples as possible to exceed a target true positive proportion (i.e.

This paper formulates model selection as an infinite-armed bandit problem. The models are arms, and picking an arm corresponds to a partial training of the model (resource allocation). The reward is the accuracy of the selected model after its partial training. In this best arm identification problem, regret is the gap between the expected accuracy of the optimal model and that of the model finally chosen. We first consider a straightforward generalization of UCB-E to the stochastic infinite-armed bandit problem and show that, under basic assumptions, the expected regret order is $T^{-\alpha}$ for some $\alpha \in (0,1/5)$ and $T$ the number of resources to allocate. From this vanilla algorithm, we introduce the algorithm Mutant-UCB that incorporates operators from evolutionary algorithms. Tests carried out on three open source image classification data sets attest to the relevance of this novel combining approach, which outperforms the state-of-the-art for a fixed budget.

We propose a new adaptive sampling approach to multiple testing which aims to maximize statistical power while ensuring anytime false discovery control. We consider $n$ distributions whose means are partitioned by whether they are below or equal to a baseline (nulls), versus above the baseline (true positives). In addition, each distribution can be sequentially and repeatedly sampled. Using techniques from multi-armed bandits, we provide an algorithm that takes as few samples as possible to exceed a target true positive proportion (i.e. Our sample complexity results match known information theoretic lower bounds and through simulations we show a substantial performance improvement over uniform sampling and an adaptive elimination style algorithm.

This paper presents an online learning approach for teams of autonomous soccer robots to select free kick plans. In robot soccer, free kicks present an opportunity to execute plans with relatively controllable initial conditions. However, the effectiveness of each plan is highly dependent on the adversary, and there are few free kicks during each game, making it necessary to learn online from sparse observations. To achieve learning, we first greatly reduce the planning space by framing the problem as a contextual multi-armed bandit problem, in which the actions are a set of pre-computed plans, and the state is the position of the free kick on the field. During execution, we model the reward function for different free kicks using Gaussian Processes, and perform online learning using the Upper Confidence Bound algorithm. Results from a physics-based simulation reveal that the robots are capable of adapting to various different realistic opponents to maximize their expected reward during free kicks.