Bounding Regret in Empirical Games
Authors: Steven Jecmen, Arunesh Sinha, Zun Li, Long Tran-Thanh4280-4287
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We present sample complexity results as well as extensive experiments that show the better performance of SAUCB compared to several baselines. |
| Researcher Affiliation | Academia | 1Carnegie Mellon University, 2Singapore Management University, 3University of Michigan, 4University of Southampton |
| Pseudocode | Yes | Super-Arm UCB (SAUCB) is our primary solution for CBBP. SAUCB is specified in Algorithm 1. |
| Open Source Code | No | The paper states 'All the full and missing proofs in this paper and additional graphical results are in the appendix of the full version, which is available on the authors webpages,' but does not mention the availability of source code for the methodology or provide a specific link. |
| Open Datasets | No | The paper uses 'synthetic data' and a 'large-scale example based on a well-known agent-based simulator of stock markets,' but it does not provide concrete access information (specific link, DOI, repository name, or formal citation with authors/year) for a publicly available or open dataset used for training. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning. |
| Hardware Specification | Yes | The run-time for each run of an algorithm in these experiments is about 6 hours (on a 2.4GHz CPU) due to the time-consuming stock market simulator, as opposed to minutes for synthetic data. |
| Software Dependencies | No | The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment. |
| Experiment Setup | Yes | We use 20 arms with Bernoulli distributions, and set W = 0.05 and α = 0.05. We examine two settings specified in this paper (the LSHN-K0 and MSMN-K0 settings with no spoofer), and bound the regret of the reported NE in each setting with α = 0.05 and W = 0.1. |