A Market for Accuracy: Classification under Competition
Ohad Einav, Nir Rosenfeld · 2025
Read A Market for Accuracy: Classification under Competition by Ohad Einav, Nir Rosenfeld (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-2502-18052.txt.
Description
Ohad Einav and Nir Rosenfeld study machine learning markets, where several providers compete for the same customers with rival models. Standard training aims only at accuracy, ignoring that a customer picks whichever provider serves her best. The authors ask what learning should look like when market share, not raw accuracy, is the prize.
They first show that accuracy alone cannot guide a provider in such a market. They then propose a classification method designed for competition, training a model to win customers in the presence of rivals. In tests from toy data to noisy real sets, the approach helped both providers and consumers, and the whole market settled quickly to a stable balance. Timing mattered: when a provider enters and how often it updates its model shaped its success.
Catalog record
| CREATOR | Ohad Einav, Nir Rosenfeld |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-18052