Tackling Interference Induced by Data Training Loops in A/B Tests: A Weighted Training Approach

N Si - arxiv preprint arxiv:2310.17496, 2023 - arxiv.org
In modern recommendation systems, the standard pipeline involves training machine
learning models on historical data to predict user behaviors and improve recommendations …

The value of external data for digital platforms: Evidence from a field experiment on search suggestions

X Lei, Y Chen, A Sen - Available at SSRN 4452804, 2023 - papers.ssrn.com
Firms increasingly leverage external data with an aim to unlock improvements in their
offerings, but it is challenging to measure the value of external data. Collaborating with a …

Random Isn't Always Fair: Candidate Set Imbalance and Exposure Inequality in Recommender Systems

A Bower, K Lum, T Lazovich, K Yee, L Belli - arxiv preprint arxiv …, 2022 - arxiv.org
Traditionally, recommender systems operate by returning a user a set of items, ranked in
order of estimated relevance to that user. In recent years, methods relying on stochastic …

Matching theory-based recommender systems in online dating

Y Tomita, R Togashi, D Moriwaki - … of the 16th ACM Conference on …, 2022 - dl.acm.org
Online dating platforms provide people with the opportunity to find a partner. Recommender
systems in online dating platforms suggest one side of users to the other side of users. This …

Experimental design for multi-channel imaging via task-driven feature selection

SB Blumberg, PJ Slator… - The Twelfth International …, 2024 - openreview.net
This paper presents a data-driven, task-specific paradigm for experimental design, to
shorten acquisition time, reduce costs, and accelerate the deployment of imaging devices …