Sample Elicitation

TL;DR

This paper proposes a deep learning-based mechanism for eliciting truthful samples from rational agents, addressing the challenge of collecting credible training data for data-intensive systems by incentivizing honest reporting.

Contribution

It introduces a novel incentive mechanism using $f$-divergence estimation to promote truthful sample collection, connecting sample elicitation with $f$-GANs for distribution reconstruction.

Findings

Mechanism effectively elicits truthful samples on synthetic and real datasets.

Theoretical guarantees for the estimator's performance.

Successful experiments on MNIST and CIFAR-10 datasets.

Abstract

It is important to collect credible training samples $(x,y)$ for building data-intensive learning systems (e.g., a deep learning system). Asking people to report complex distribution $p(x)$, though theoretically viable, is challenging in practice. This is primarily due to the cognitive loads required for human agents to form the report of this highly complicated information. While classical elicitation mechanisms apply to eliciting a complex and generative (and continuous) distribution $p(x)$, we are interested in eliciting samples $x_i \sim p(x)$ from agents directly. We coin the above problem "sample elicitation". This paper introduces a deep learning aided method to incentivize credible sample contributions from self-interested and rational agents. We show that with an accurate estimation of a certain $f$-divergence function we can achieve approximate incentive compatibility in eliciting truthful samples. We then present an efficient estimator with theoretical guarantees via studying the variational forms of the $f$-divergence function. We also show a connection between this sample elicitation problem and $f$-GAN, and how this connection can help reconstruct an estimator of the distribution based on collected samples. Experiments on synthetic data, MNIST, and CIFAR-10 datasets demonstrate that our mechanism elicits truthful samples. Our implementation is available at https://github.com/weijiaheng/Credible-sample-elicitation.git.