XSci

IncVNM: An Interactive Recommender System with Incremental von Neumann–Morgenstern Utility Learning

Jiaxin Wei, Jia Liu

Submitted by Jiaxin Wei

Published November 22, 2026 · Version v1, November 22, 2026 · DOI 10.66977/xsci.2611.003w

Human-Computer Interaction, Algorithms, Applied Mathematics

Abstract

In modern recommender systems, users are often presented with products whose outcomes are random. The values of such products depend on their entire outcome distributions and on users’ attitudes toward risk, making them difficult to evaluate using traditional product-level recommendation methods. We propose IncVNM, an interactive recommender system that incrementally learns a userspecific von Neumann–Morgenstern (VNM) utility function from ordinal click feedback generated during ordinary webpage browsing. At each interaction, IncVNM converts the click order within a displayed product list into pairwise comparisons between outcome distributions and updates the nominal utility by minimizing its deviation from the previous estimate while incorporating the newly observed pairwise comparisons. To close the interaction loop, an entropy-regularized sampling policy balances exploitation of the learned utility against exploration for additional preference information. We evaluate IncVNM in a rolling-window experiment with different response-error probabilities and show that it substantially outperforms the expected-return, mean–variance, discounted-UCB, and distribution-aware BPR baselines.

View PDF