Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

Imprecise Multi-Armed Bandits: Representing Irreducible Uncertainty as a Zero-Sum Game

Authors: Vanessa Kosoy

JMLR 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We introduce a novel multi-armed bandit framework, where each arm is associated with a fixed unknown credal set over the space of outcomes... We then define a notion of regret corresponding to the lower prevision defined by these credal sets... For certain natural hypothesis classes... we propose an algorithm and prove a corresponding upper bound on regret. Keywords: multi-armed bandits, game theory, regret bounds, upper confidence bound, imprecise probability
Researcher Affiliation Academia Vanessa Kosoy EMAIL Faculty of Mathematics Technion Israel Institute of Technology Haifa 3200003, Israel. Association for Long Term Existence and Resilience Rehovot, Israel.
Pseudocode Yes Algorithm 1 Imprecise UCB Input η R+ C H for k from 1 to : θ argmaxθ C maxx A MEθ [r|x] τ 0 Σy 0 Y do select arm x θ and observe outcome y τ τ + 1 Σy Σy + y y Σy while τ maxθ C minz V(x θ , y) θ z < 2 (DZ + 1) η C n θ C minz V(x θ , y) θ z η τ
Open Source Code No The paper does not contain any explicit statements or links indicating that source code for the described methodology is publicly available.
Open Datasets No The paper uses illustrative examples (e.g., traffic lights, patient treatment) to explain the framework but does not perform experiments with specific datasets, nor does it provide access information for any datasets.
Dataset Splits No The paper is theoretical and does not describe experiments using datasets, therefore, there is no mention of dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and focuses on developing a new framework and algorithm with mathematical proofs. It does not describe any computational experiments or hardware used.
Software Dependencies No The paper is theoretical and does not describe any implementation or experimental setup, thus no specific software dependencies or version numbers are mentioned.
Experiment Setup No The paper is theoretical, presenting a new framework and an algorithm. It does not describe any experimental setup details such as hyperparameters, training configurations, or system-level settings.