EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3028. Contributions of surrogate models and counterfactual explanations to marketplace blackbox audits

Invited abstract in session TA-27: Counterfactual Analysis Across Diverse Domains, stream Mathematical Optimization for XAI.

Tuesday, 8:30-10:00
Room: 047 (building: 208)

Authors (first author is the speaker)

1. Benoit Rottembourg
Regalia, INRIA
2. Jeanne Mouton
Université Nice Côte d'Azur
3. Jordan Thieyre

Abstract

In a global context where competition authorities are investigating and sanctioning Amazon marketplace for practices of self-preferencing at the expense of their business users and consumers, we observe a trend of imposing remedies on dominant players in digital markets.
Therefore, competition authorities and regulators need tools to audit the compliance of these dominant players in the e-commerce sector over the obligations and remedies they are imposing on dynamic, and personalized algorithms. Most of these algorithms embed Machine-Learning components, introducing opacity and potentially biases in the decision-making process.

The aim of our presentation is to explore the benefits of using black-box auditing techniques and counterfactual explanations to provide insights into the behavior of these online algorithms. We anchor our research in the literature of product preeminence from vertically integrated players, of choice ranking, and of the specific literature related to Amazon search ranking, automatic pricing and Buy Box's algorithms ([1];[2]).
Through a longitudinal study of the ranking of several thousand products on Amazon, we will illustrate the potential of "surrogate" models and the decision-support elements they might provide.

[1] Chen et al "An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace" 2016
[2] Gómez-Losada et al "Automatic Eligibility of Sellers in an Online Marketplace: A Case Study of Amazon Algorithm" 2022

Keywords

Status: accepted

Back to the list of papers