Subjective Logic Encodings

TL;DR

This paper introduces Subjective Logic Encodings (SLEs), a novel framework that explicitly encodes annotator opinions and uncertainties in labels, enabling models to better handle subjective tasks with inherent disagreement.

Contribution

The paper proposes SLEs, a flexible, principled method for encoding and aggregating annotation uncertainty based on Subjective Logic Theory, expanding data perspectivism beyond disagreement as the sole uncertainty source.

Findings

SLEs encode labels as Dirichlet distributions.

SLEs generalize existing label encoding methods.

Models can predict SLEs using distribution matching.

Abstract

Many existing approaches for learning from labeled data assume the existence of gold-standard labels. According to these approaches, inter-annotator disagreement is seen as noise to be removed, either through refinement of annotation guidelines, label adjudication, or label filtering. However, annotator disagreement can rarely be totally eradicated, especially on more subjective tasks such as sentiment analysis or hate speech detection where disagreement is natural. Therefore, a new approach to learning from labeled data, called data perspectivism, seeks to leverage inter-annotator disagreement to learn models that stay true to the inherent uncertainty of the task by treating annotations as opinions of the annotators, rather than gold-standard facts. Despite this conceptual grounding, existing methods under data perspectivism are limited to using disagreement as the sole source of annotation uncertainty. To expand the possibilities of data perspectivism, we introduce Subjective Logic Encodings (SLEs), a flexible framework for constructing classification targets that explicitly encodes annotations as opinions of the annotators. Based on Subjective Logic Theory, SLEs encode labels as Dirichlet distributions and provide principled methods for encoding and aggregating various types of annotation uncertainty -- annotator confidence, reliability, and disagreement -- into the targets. We show that SLEs are a generalization of other types of label encodings as well as how to estimate models to predict SLEs using a distribution matching objective.