Multi-sensor Information Processing using Prediction Market-based Belief Aggregation

TL;DR

This paper introduces a novel multi-agent prediction market approach for multi-sensor data fusion, improving inference confidence and classification accuracy in landmine detection by incentivizing truthful reporting.

Contribution

The paper presents a new distributed belief aggregation method using prediction markets with a market maker, enhancing sensor data fusion and incentivizing truthful reporting.

Findings

Improved classification accuracy over traditional methods.

Effective incentive scheme for truthful sensor reporting.

Validated approach in landmine detection scenario.

Abstract

We consider the problem of information fusion from multiple sensors of different types with the objective of improving the confidence of inference tasks, such as object classification, performed from the data collected by the sensors. We propose a novel technique based on distributed belief aggregation using a multi-agent prediction market to solve this information fusion problem. To monitor the improvement in the confidence of the object classification as well as to dis-incentivize agents from misreporting information, we have introduced a market maker that rewards the agents instantaneously as well as at the end of the inference task, based on the quality of the submitted reports. We have implemented the market maker's reward calculation in the form of a scoring rule and have shown analytically that it incentivizes truthful revelation or accurate reporting by each agent. We have experimentally verified our technique for multi-sensor information fusion for an automated landmine detection scenario. Our experimental results show that, for identical data distributions and settings, using our information aggregation technique increases the accuracy of object classification favorably as compared to two other commonly used techniques for information fusion for landmine detection.