Robust Multi-agent Counterfactual Prediction

Authors: Alexander Peysakhovich, Christian Kroer, Adam Lerer

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We apply RMAC to classic environments in market design: auctions, school choice, and social choice. and 6 Experiments We now turn to constructing RMAC bounds for classic problems in market design.
Researcher Affiliation Industry Alexander Peysakhovich Facebook AI Research Christian Kroer Facebook Core Data Science Adam Lerer Facebook AI Research
Pseudocode Yes Algorithm 1 Revelation Fictitious Play
Open Source Code No The paper does not provide any explicit statement about releasing source code or a link to a code repository.
Open Datasets No We generate data by first sampling 1000 independent types and their actions from the closed form first-price equilibrium (bid = .5θ), using these actions as D. This indicates synthetic data generation, not the use of a public dataset with access information.
Dataset Splits No The paper describes generating synthetic data and using it as 'D', but does not specify any explicit train/validation/test splits for this data.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, memory, or cloud instance types used for experiments.
Software Dependencies No The paper does not list any specific software dependencies with version numbers.
Experiment Setup Yes We consider a first-price 2-player auction G with types drawn from [0, 1] uniformly and bids in the interval [0, 1] discretized at intervals of .01. ... We set the domain of possible types to also be equal to [0, 1]. We generate data by first sampling 1000 independent types and their actions from the closed form first-price equilibrium (bid = .5θ), using these actions as D. We then use D to compute ϵ-RMAC predictions for several levels of ϵ.