Year: 2023
Type: preprint
Abstract: A key puzzle in search, ads, and recommendation is that the ranking model can only utilize a small portion of the vastly available user interaction data. As a result, increasing data volume, model size, or computation FLOPs will quickly suffer from diminishing returns. We examined this problem and found that one of the root causes may lie in the so-called ``item-centric'' formulation, which has an unbounded vocabulary and thus uncontrolled model complexity. To mitigate quality saturation, we introduce an alternative formulation named ``user-centric ranking'', which is based on a transposed view of the dyadic user-item interaction data. We show that this formulation has a promising scaling property, enabling us to train better-converged models on substantially larger data sets.
Source: arXiv (Cornell University)
Authors: Zhao, Zhuokai, Yang, Yang, Wang, Wenyu, Liu, Chihuang, Shi, Yu, Hu, Wenjie, Zhang, Haotian, Yang, Shuang
Institutions: University of Chicago, Meta (United States)
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4378474264