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.
Locally Private and Robust Multi-Armed Bandits
Authors: Xingyu Zhou, Komo(Wei) ZHANG
NeurIPS 2024 | Venue PDF | LLM Run Details | Input Tokens: 36,260 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,819 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 | Beyond our theoretical results, we have also conducted a set of simulations for our three problems. |
| Researcher Affiliation | Academia | Xingyu Zhou Wayne State University EMAIL Wei Zhang Texas A&M University EMAIL |
| Pseudocode | Yes | Algorithm 1 A Unified Algorithm |
| Open Source Code | Yes | We provide the code with detailed instructions for the experiments discussed in Appendix A. |
| Open Datasets | No | We consider Pareto distribution, whose probability distribution is given by f(x; xm, s) = sxs m xs+1 , if x xm 0, otherwise where s > 0 is the shape parameter and xm > 0 is the scale parameter. |
| Dataset Splits | No | The paper describes generating synthetic data based on a Pareto distribution for simulations but does not specify any training, validation, or test dataset splits. |
| Hardware Specification | No | Our experiments are designed primarily to support the theoretical results and are relatively simple in their settings. They do not require high-performance hardware and can be run on most standard computers. |
| Software Dependencies | No | The paper does not explicitly list specific software dependencies with version numbers for reproducibility. |
| Experiment Setup | Yes | In our experiments, we choose k = 2 and consider various corruption level α {0, 0.02, 0.05} and privacy budget ε {0.3, 0.5, 1}. For each set of parameters, we conduct 300 runs and plot the average of the estimation error and corresponding confidence region. |