A neuro-symbolic benchmark suite for concept quality and reasoning shortcuts

Samuele Bortolotti, Emanuele Marconato, Tommaso Carraro, Paolo Morettin, Emile van Krieken, Antonio Vergari, Stefano Teso, Andrea Passerini

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The advent of powerful neural classifiers has increased interest in problems that require both learning and reasoning. These problems are critical for understanding important properties of models, such as trustworthiness, generalization, interpretability, and compliance to safety and structural constraints. However, recent research observed that tasks requiring both learning and reasoning on background knowledge often suffer from reasoning shortcuts (RSs): predictors can solve the downstream reasoning task without associating the correct concepts to the high-dimensional data. To address this issue, we introduce rsbench, a comprehensive benchmark suite designed to systematically evaluate the impact of RSs on models by providing easy access to highly customizable tasks affected by RSs. Furthermore, rsbench implements common metrics for evaluating concept quality and introduces novel formal verification procedures for assessing the presence of RSs in learning tasks. Using rsbench, we highlight that obtaining high quality concepts in both purely neural and neuro-symbolic models is a far-from-solved problem. rsbench is available at: https://unitn-sml.github.io/rsbench.
Original languageEnglish
Title of host publicationProceedings of the Thirty-Eighth Annual Conference on Neural Information Processing Systems
PublisherACM
Pages1-44
Number of pages44
DOIs
Publication statusAccepted/In press - 26 Sept 2024
EventThe Thirty-Eighth Annual Conference on Neural Information Processing Systems - Vancouver Convention Center, Vancouver, Canada
Duration: 10 Dec 202415 Dec 2024
Conference number: 38
https://neurips.cc/Conferences/2024

Conference

ConferenceThe Thirty-Eighth Annual Conference on Neural Information Processing Systems
Abbreviated titleNeurIPS 2024
Country/TerritoryCanada
CityVancouver
Period10/12/2415/12/24
Internet address

Keywords / Materials (for Non-textual outputs)

  • machine learning
  • artificial intelligence

Fingerprint

Dive into the research topics of 'A neuro-symbolic benchmark suite for concept quality and reasoning shortcuts'. Together they form a unique fingerprint.

Cite this