Inducing Equilibria via Incentives: Simultaneous Design-and-Play Ensures Global Convergence

Authors: Boyi Liu, Jiayang Li, Zhuoran Yang, Hoi-To Wai, Mingyi Hong, Yu Nie, Zhaoran Wang

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we conduct two numerical experiments to test our algorithms. All numerical results reported in this section were produced either on a Mac Book Pro (15-inch, 2017) with 2.9 GHz Quad-Core Intel Core i7 CPU.
Researcher Affiliation Academia Boyi Liu , Jiayang Li Northwestern University {boyiliu2018,jiayangli2024}@u.northwestern.edu Zhuoran Yang Yale University zhuoran.yang@yale.edu Hoi-To Wai The Chinese University of Hong Kong htwai@se.cuhk.edu.hk Mingyi Hong University of Minnesota mhong@umn.edu Yu (Marco) Nie, Zhaoran Wang Northwestern University {y-nie,zhaoran.wang}@northwestern.edu
Pseudocode Yes Algorithm 1 Bilevel incentive design for unconstrained games
Open Source Code No The paper does not provide any statement or link indicating that the source code for the methodology is openly available.
Open Datasets Yes We test our algorithm on a real-world traffic network: the Sioux-Falls network (See Lawphongpanich and Hearn [24] for its structure).
Dataset Splits No The paper does not explicitly mention training, validation, and test dataset splits with specific percentages or counts. It describes the overall experimental setup and data used but not the partitioning for model development.
Hardware Specification Yes All numerical results reported in this section were produced either on a Mac Book Pro (15-inch, 2017) with 2.9 GHz Quad-Core Intel Core i7 CPU.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies. It mentions implementing in Python, but no specific versions for Python or any libraries/frameworks.
Experiment Setup Yes To make a fair comparison, the same hyperparameters including the initial solutions, the learning rates, and the tolerance values for both upperand lower-level problems are employed for the tested algorithms (double-loop AD, double-loop ID, and our algorithm). ... Setting A: k = /(k + 1)1/2, βk = β/(k + 1)2/7, and k = /(k + 1)4/7.