Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
An Equivalence between Wagering and Fair-Division Mechanisms
Authors: Rupert Freeman, David M. Pennock, Jennifer Wortman Vaughan1957-1964
AAAI 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We next compare the performance of these mechanisms in practice. Our test data comes from a binary outcome (m = 2) wagering setting, but the equivalence allows us to make observations about allocation-specific properties like envy-freeness too. In the full version of the paper, we additionally report results from simulations with larger outcome spaces using synthetic data. |
| Researcher Affiliation | Industry | Rupert Freeman Microsoft Research EMAIL David M. Pennock Microsoft Research EMAIL Jennifer Wortman Vaughan Microsoft Research EMAIL |
| Pseudocode | No | The paper describes various mechanisms and their properties through definitions and textual explanations, but it does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing its source code or a link to a code repository for the described methodology. |
| Open Datasets | Yes | We use reports gathered from an online prediction contest called Probability Sports (Galebach 2017). 2The dataset consists of probabilistic predictions about the outcomes of 1643 U.S. National Football League matches between 2000 and 2004. |
| Dataset Splits | No | The paper uses the 'Probability Sports' dataset and describes its characteristics, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts) for reproduction. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory used for running its experiments. |
| Software Dependencies | No | The paper refers to concepts like 'Brier score' and various mechanisms, but it does not provide specific software dependency details with version numbers (e.g., Python, PyTorch, or specific solver versions) needed to replicate the experiments. |
| Experiment Setup | Yes | For our simulations, we follow previous work (Freeman, Pennock, and Vaughan 2017) and generate wagers in two ways. First, we consider uniform wagers, modeling the scenario in which agents are required to risk the same amount. Second, for smaller instances on which it is tractable to employ the agent duplication method for unequal weights, we generate wagers according to a Pareto distribution with shape parameter 1.16 and scale parameter 1. Since agent duplication requires rational wagers, we scale the generated wagers to lie in [0, 50], and then take the ceiling of each, yielding integral wagers between 1 and 50. |