On Simple Mechanisms for Dependent Items

TL;DR

This paper investigates simple selling mechanisms for a single buyer with dependent valuations modeled by Markov Random Fields, providing approximation guarantees that depend on the dependency structure and extending known results to complex valuation classes.

Contribution

It extends revenue approximation results to dependent item valuations modeled by MRFs, introducing new bounds and techniques for XOS and other valuation classes.

Findings

Max(SRev,BRev) achieves an exponential approximation depending on MRF parameter Δ.

Results recover constant-factor approximations in the independent case.

New concentration inequality for XOS functions over dependent variables.

Abstract

We study the problem of selling $n$ heterogeneous items to a single buyer, whose values for different items are dependent. Under arbitrary dependence, Hart and Nisan show that no simple mechanism can achieve a non-negligible fraction of the optimal revenue even with only two items. We consider the setting where the buyer's type is drawn from a correlated distribution that can be captured by a Markov Random Field, one of the most prominent frameworks for modeling high-dimensional distributions with structure. If the buyer's valuation is additive or unit-demand, we extend the result to all MRFs and show that max(SRev,BRev) can achieve an $\Omega\left(\frac{1}{e^{O(\Delta)}}\right)$-fraction of the optimal revenue, where $\Delta$ is a parameter of the MRF that is determined by how much the value of an item can be influenced by the values of the other items. We further show that the exponential dependence on $\Delta$ is unavoidable for our approach and a polynomial dependence on $\Delta$ is unavoidable for any approach. When the buyer has a XOS valuation, we show that max(Srev,Brev) achieves at least an $\Omega\left(\frac{1}{e^{O(\Delta)}+\frac{1}{\sqrt{n\gamma}}}\right)$-fraction of the optimal revenue, where $\gamma$ is the spectral gap of the Glauber dynamics of the MRF. Note that in the special case of independently distributed items, $\Delta=0$ and $\frac{1}{n\gamma}\leq 1$, and our results recover the known constant factor approximations for a XOS buyer. We further extend our parametric approximation to several other well-studied dependency measures such as the Dobrushin coefficient and the inverse temperature. Our results are based on the Duality-Framework by Cai et al. and a new concentration inequality for XOS functions over dependent random variables.