Federated X-armed Bandit
Authors: Wenjie Li, Qifan Song, Jean Honorio, Guang Lin
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments We empirically evaluate the proposed Fed-PNE algorithm on both synthetic functions and real-world datasets. We compare Fed-PNE with centralized X-armed bandit algorithm HCT (Azar, Lazaric, and Brunskill 2014), centralized kernelized bandit algorithm Kernel TS (Chowdhury and Gopalan 2017), federated multi-armed bandit algorithm Fed1-UCB (Shi and Shen 2021a), and federated neural bandit algorithm FN-UCB (Dai et al. 2023). ...The average cumulative regret of different algorithms are provided in Figure 2(a) and 2(b). |
| Researcher Affiliation | Academia | Wenjie Li1*, Qifan Song1, Jean Honorio2, Guang Lin 3 1Department of Statistics, Purdue University 2School of Computing and Information Systems, The University of Melbourne 3Departments of Mathematics and School of Mechanical Engineering, Purdue University |
| Pseudocode | Yes | Algorithm 1: Fed-PNE: m-th client and Algorithm 2: Fed-PNE: server |
| Open Source Code | No | The paper cites 'Li, W.; Li, H.; Honorio, J.; and Song, Q. 2023a. Py XAB A Python Library for X-Armed Bandit and Online Blackbox Optimization Algorithms.' in its references, but does not explicitly state that the source code for the Fed-PNE algorithm described in this specific paper is available or link directly to it. |
| Open Datasets | Yes | Landmine Detection. We federatedly tune the hyperparameters of machine learning models fitted on the Landmine dataset (Liu, Liao, and Carin 2007) and COVID-19 Vaccine Dosage Optimization. In combat to the pandemic, we optimize the vaccine dosage in epidemiological models of COVID-19 to find the best fractional dosage for the overall population following Wiecek et al. (2022). |
| Dataset Splits | No | The paper describes the datasets used and how data is distributed among clients (e.g., 'each client only has the access to the data of one random field'), but it does not specify exact train/validation/test split percentages, sample counts, or refer to predefined splits with citations for reproduction. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to conduct the experiments. |
| Software Dependencies | No | The paper mentions various algorithms and libraries (e.g., 'Py XAB A Python Library for X-Armed Bandit'), but it does not provide specific software dependencies with version numbers (e.g., 'Python 3.8', 'TensorFlow 2.x') required to reproduce the experiments. |
| Experiment Setup | Yes | Landmine Detection... trains a support vector machine with the RBF kernel parameter chosen from [0.01, 10] and the L2 regularization parameter chosen from [10 4, 10]. |