Direct Profit Estimation Using Uplift Modeling under Clustered Network Interference Conference: The 19th ACM Conference on Recommender Systems; September 22, 2025; Prague, Czech Republic

Bram van den Akker · 2025

Read Direct Profit Estimation Using Uplift Modeling under Clustered Network Interference Conference: The 19th ACM Conference on Recommender Systems; September 22, 2025; Prague, Czech Republic by Bram van den Akker (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-01558.txt.

Description

Promotions in online marketplaces do not act on one item alone: treating one product can change the fate of its neighbors. Bram van den Akker studies uplift modeling, a method that estimates the extra profit a promotion causes, under such clustered network interference. Standard uplift methods assume no interference, an assumption whose failure leads to poor policies in real marketplaces.

Van den Akker builds a practical method around an interference-aware estimator, used as a learning objective that gradient methods can optimize directly. Combined with proven response transformation techniques, the framework aims straight at economic outcomes such as incremental profit. Simulations show the approach beats interference-naive methods by a wide margin, above all when interference effects grow. Guidance is also given for adapting profit estimation to settings where items influence each other in clusters.

Catalog record

CREATORBram van den Akker
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE17 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-01558

Browse the full library · About the Tower of Babel