Minimum Weighted Feedback Arc Sets for Ranking from Pairwise Comparisons
Abstract
Combinatorial algorithms for the Minimum Weighted Feedback Arc Set problem provide fast, learning-free alternatives that match or exceed deep learning methods on large-scale ranking tasks.
The Minimum Weighted Feedback Arc Set (MWFAS) problem is closely related to the task of deriving a global ranking from pairwise comparisons. Recent work by He et al. (ICML 2022) advanced the state of the art on ranking benchmarks using learning based methods, but did not examine the underlying connection to MWFAS. In this paper, we investigate this relationship and introduce efficient combinatorial algorithms for solving MWFAS as a means of addressing the ranking problem. Our experimental results show that these simple, learning free methods achieve substantially faster runtimes than recent learning based approaches, while also delivering competitive, and in many cases superior, ranking accuracy. These findings suggest that lightweight combinatorial techniques offer a scalable and effective alternative to deep learning for large scale ranking tasks.
Get this paper in your agent:
hf papers read 2412.16181 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 2
SoroushVahidi/ranking-fas-results
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper