English

Loss Functions, Axioms, and Peer Review

Artificial Intelligence 2020-03-03 v2 Computer Science and Game Theory Machine Learning

Abstract

It is common to see a handful of reviewers reject a highly novel paper, because they view, say, extensive experiments as far more important than novelty, whereas the community as a whole would have embraced the paper. More generally, the disparate mapping of criteria scores to final recommendations by different reviewers is a major source of inconsistency in peer review. In this paper we present a framework inspired by empirical risk minimization (ERM) for learning the community's aggregate mapping. The key challenge that arises is the specification of a loss function for ERM. We consider the class of L(p,q)L(p,q) loss functions, which is a matrix-extension of the standard class of LpL_p losses on vectors; here the choice of the loss function amounts to choosing the hyperparameters p,q[1,]p, q \in [1,\infty]. To deal with the absence of ground truth in our problem, we instead draw on computational social choice to identify desirable values of the hyperparameters pp and qq. Specifically, we characterize p=q=1p=q=1 as the only choice of these hyperparameters that satisfies three natural axiomatic properties. Finally, we implement and apply our approach to reviews from IJCAI 2017.

Keywords

Cite

@article{arxiv.1808.09057,
  title  = {Loss Functions, Axioms, and Peer Review},
  author = {Ritesh Noothigattu and Nihar B. Shah and Ariel D. Procaccia},
  journal= {arXiv preprint arXiv:1808.09057},
  year   = {2020}
}
R2 v1 2026-06-23T03:45:26.221Z