Crowdsourced Outcome Determination in Prediction Markets

Authors: Rupert Freeman, Sebastien Lahaie, David Pennock

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we propose a specific prediction market mechanism with crowdsourced outcome determination that addresses several challenges faced by decentralized markets of this sort. Additionally, by allowing anyone to create a market, there is no longer any guarantee that all questions will be sensible, or even have a well-defined outcome. Our analysis will assume that μ1 and μ0 are common across agents, but this is not a strict requirement. If we allow each agent to have distinct updates μi 1, μi 0, then we can let μ1 = mini μi 1, corresponding to the minimum update given ˆxi = 1, and similarly μ0 = maxi μi 0. In this section, we derive conditions for truthful reporting (ˆxi = xi) to be a best response, given that all other arbiters report truthfully. We plot (3) and (4) in Figure 2, considering their tight versions as equalities. This paper proposed and analyzed a mechanism where the outcome of an MSR prediction market is determined via a peer prediction mechanism among a set of arbiters.
Researcher Affiliation Collaboration Rupert Freeman Duke University rupert@cs.duke.edu S ebastien Lahaie Microsoft Research slahaie@microsoft.com David M. Pennock Microsoft Research dpennock@microsoft.com
Pseudocode Yes Figure 1: Prediction market with outcome determined using peer prediction. 1. Market stage. (a) A prediction market is set up for an event X using a market scoring rule. (b) Agents trade in the market. For a security bought at price p, a trading fee of fp is charged, and for a security sold at price p, a fee of f(1 p) is charged. (c) The market closes, trading stops. 2. Arbitration stage. (a) Each arbiter i receives a signal xi {0, 1} and reports an outcome ˆxi {0, 1}. (b) Each arbiter i is assigned a peer arbiter j = i and paid according to the 1/prior with midpoint mechanism. (c) The outcome of the market, and the payoff of each share sold, is set to the fraction of arbiters that report ˆxi = 1.
Open Source Code No The paper does not provide any statement or link indicating that its source code is publicly available.
Open Datasets No The paper is theoretical and does not describe experiments using any specific dataset, public or otherwise, nor does it provide any access information or citations for data.
Dataset Splits No The paper does not describe any experimental setup involving datasets, so there is no mention of training, validation, or test splits.
Hardware Specification No The paper is theoretical and does not describe any experiments that would require specific hardware, therefore no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not describe any experiments that would require specific software with version numbers, therefore no software dependencies are mentioned.
Experiment Setup No The paper is theoretical and focuses on mechanism design and analysis. It does not describe an experimental setup with hyperparameters or training configurations.