Bayesian Persuasion in Sequential Decision-Making
Authors: Jiarui Gan, Rupak Majumdar, Goran Radanovic, Adish Singla5025-5033
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically evaluate signaling strategies obtained with our algorithms. The goal is to compare the payoffs yielded for the principal. We use Python (v3.9) to implement our algorithms and Gurobi (v9.1.2) to solve all the LPs. All results were obtained on a platform with a 2 GHz Quad-Core CPU and 16 GB memory, and each is averaged over at least 20 instances. We conduct experiments on (i) general instances without any specific underlying structure, and (ii) instances generated based on a road navigation application. |
| Researcher Affiliation | Academia | Jiarui Gan, Rupak Majumdar, Goran Radanovic, Adish Singla Max Planck Institute for Software Systems {jrgan,rupak,gradanovic,adishs}@mpi-sws.org |
| Pseudocode | No | No pseudocode or clearly labeled algorithm blocks were found in the paper. |
| Open Source Code | No | The paper does not provide any statement or link indicating that the source code for the described methodology is open-source or publicly available. |
| Open Datasets | No | The paper describes generating its own instances for experiments ('general instances', 'road navigation application') but does not provide any access information (link, DOI, citation) for a publicly available or open dataset. |
| Dataset Splits | No | The paper does not explicitly provide specific training/test/validation dataset splits or cross-validation details needed for reproduction. |
| Hardware Specification | Yes | All results were obtained on a platform with a 2 GHz Quad-Core CPU and 16 GB memory |
| Software Dependencies | Yes | We use Python (v3.9) to implement our algorithms and Gurobi (v9.1.2) to solve all the LPs. |
| Experiment Setup | Yes | In all figures, we fix |S| = |Θ| = |A| = 10, γ = γ = 0.8, n = 5, and β = 0 unless they are variables. ... All results are obtained on instances with n = 20 and m = 100 (i.e., numbers of nodes and edges in the network), where we also fix |Θ| = 3, γ = γ = 0.8, and β = 0.5 unless they are variables. |