The paper introduces SHPRR, an algorithm that adapts Sequential Halving for pure exploration problems where regret relative to a baseline is a key cost metric. It shows that regret cost decreases as the algorithm progresses, allowing a front‑loaded allocation of the regret budget to improve best‑arm identification. Experiments on synthetic and real data demonstrate that SHPRR outperforms uniform regret allocation and existing baselines.
By Georgina Nouli, Flore Sentenac, Maximilian Schiffer
arXiv:2607. 28408v1 Announce Type: new Abstract: This thesis studies policy learning in interactive systems where an agent observes a context, selects an action from a very large set, and receives partial feedback.
By Imad Aouali
The paper introduces TS-SA, a variant of Langevin-based Thompson Sampling that integrates stochastic approximation into the bandit framework. TS-SA performs a single Langevin step per new reward and updates a persistent Markov chain for each arm, avoiding repeated sampling of past data. This approach yields a near‑optimal worst‑case regret bound of ≥O(√{KT}) under log‑concavity and smoothness, and empirical results show strong performance against standard baselines.
By Weixin Wang, Haoyang Zheng, Guang Lin, Wei Deng, Pan Xu
arXiv:2606. 01655v1 Announce Type: cross Abstract: The Bayesian paradigm offers principled tools for sequential decision-making under uncertainty, but its reliance on a probabilistic model for all parameters can hinder the incorporation of complex structural constraints.
By Kaizheng Wang
arXiv:2606. 26657v1 Announce Type: new Abstract: Identifying high-utility candidates from massive discrete spaces under expensive evaluations is a recurring challenge across the sciences, with structure-based drug discovery as a prominent example.
By Mohammad Haddadnia, Yuvan Chali, Abhilash Jayaraj, Constance Kraay, Joana Reis, Felix Strieth-Kalthoff, Haribabu Arthanari
arXiv:2606. 28616v1 Announce Type: new Abstract: In stochastic linear bandits, the canonical Upper Confidence Bound (UCB) algorithm admits a simple frequentist regret analysis but can be computationally demanding, while Thompson Sampling (TS) is computationally attractive yet typically harder to analyze due to its non-optimistic nature.
By Toshinori Kitamura, Shuai Liu, Csaba Szepesv\'ari
arXiv:2502. 01226v4 Announce Type: replace Abstract: Gaussian process (GP) bandits provide a powerful framework for performing blackbox optimization of unknown functions.
By Jack Sandberg, Morteza Haghir Chehreghani
arXiv:2506. 24007v5 Announce Type: replace-cross Abstract: This study investigates minimax and Bayes optimal strategies for fixed-budget best-arm identification.
By Masahiro Kato
arXiv:2605. 20854v2 Announce Type: replace Abstract: We study a stochastic bandit algorithm motivated by retry-aware objectives that value the best outcome among multiple attempts, such as pass@$k$ and max@$k$.
By Bingkui Tong, Junpei Komiyama, Soichiro Nishimori, Paavo Parmas
arXiv:2602. 17976v2 Announce Type: replace-cross Abstract: In active sequential testing, also termed pure exploration, a learner is tasked with the goal to adaptively acquire information so as to identify an unknown ground-truth hypothesis with as few queries as possible.
By Alessio Russo, Yin-Ching Lee, Ryan Welch, Aldo Pacchiano
The paper introduces a method that integrates action abstraction into policy optimization for reinforcement learning and generative flow networks. By iteratively identifying frequently used action subsequences in high‑reward trajectories and treating them as single high‑level actions, the approach expands the action space and improves sample efficiency. Experiments on synthetic and real‑world tasks show that this technique discovers diverse high‑reward states more effectively, especially on challenging exploration problems, and yields interpretable abstract actions that reflect the underlying reward structure.
By Oussama Boussif, L\'ena N\'ehale Ezzine, Joseph D Viviano, Micha{\l} Koziarski, Moksh Jain, Esmeralda S. Whitammer, Emmanuel Bengio, Rim Assouel, Yoshua Bengio
In-Context Batch Black-box Optimization via Flow Matching introduces FMTR, an amortized method that learns a continuous, history-conditioned proposal distribution for variable-size batch sampling in expensive black-box optimization settings. FMTR is trained offline using conditional flow matching on Thompson-sampling argmaxes refined through a horizon-dependent Langevin pushforward map, enabling adaptive exploration-exploitation over the optimization horizon. At test time, FMTR generates batches without task-specific surrogate refitting or numerical acquisition optimization, achieving strong performance on synthetic and real-world benchmarks while significantly reducing batch selection computational cost.
By Quoc-Anh Hoang Nguyen, Xinyu Zhang, Daolang Huang, Julien Martinelli, Samuel Kaski