Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
Optimal Estimation of the Best Mean in Multi-Armed Bandits
Authors: Takayuki Osogami, Junya Honda, Junpei Komiyama
NeurIPS 2025 | Venue PDF | LLM Run Details | Input Tokens: 37,322 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 1,516 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results support our theoretical guarantees and demonstrate the practical effectiveness of our method. |
| Researcher Affiliation | Collaboration | Takayuki Osogami IBM Research Tokyo EMAIL Junya Honda Kyoto University, RIKEN AIP EMAIL Junpei Komiyama New York University, RIKEN AIP EMAIL |
| Pseudocode | Yes | Algorithm 1 Ellipsoid Est |
| Open Source Code | Yes | All the details of experimental settings are explained in Section 7 or Appendix B, and source code is submitted as the supplementary material. |
| Open Datasets | No | This paper does not use any datasets. |
| Dataset Splits | No | Our experimental settings do not involve train-test split. |
| Hardware Specification | Yes | All experiments in Appendix B, including those reported previously, were conducted on a single CPU core with 4 GB of memory and no GPU acceleration, in a cloud environment. |
| Software Dependencies | Yes | We use numpy, scipy, matplotlib, jupyterlab, which are explicitly mentioned in pyproject.toml files. |
| Experiment Setup | Yes | We choose to set the regularization parameter as λ = (R/S)2. ... We fix ε = 0.1 and study the impact of varying δ (the confidence level) and K (the number of arms). |