Not All Neuro-Symbolic Concepts Are Created Equal: Analysis and Mitigation of Reasoning Shortcuts

05/31/2023
by   Emanuele Marconato, et al.
0

Neuro-Symbolic (NeSy) predictive models hold the promise of improved compliance with given constraints, systematic generalization, and interpretability, as they allow to infer labels that are consistent with some prior knowledge by reasoning over high-level concepts extracted from sub-symbolic inputs. It was recently shown that NeSy predictors are affected by reasoning shortcuts: they can attain high accuracy but by leveraging concepts with unintended semantics, thus coming short of their promised advantages. Yet, a systematic characterization of reasoning shortcuts and of potential mitigation strategies is missing. This work fills this gap by characterizing them as unintended optima of the learning objective and identifying four key conditions behind their occurrence. Based on this, we derive several natural mitigation strategies, and analyze their efficacy both theoretically and empirically. Our analysis shows reasoning shortcuts are difficult to deal with, casting doubts on the trustworthiness and interpretability of existing NeSy solutions.

READ FULL TEXT

page 9

page 26

research
03/22/2023

Neuro-Symbolic Reasoning Shortcuts: Mitigation Strategies and their Limitations

Neuro-symbolic predictors learn a mapping from sub-symbolic inputs to hi...
research
02/02/2023

Neuro Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal

We introduce Neuro-Symbolic Continual Learning, where a model has to sol...
research
02/16/2023

Empirical Investigation of Neural Symbolic Reasoning Strategies

Neural reasoning accuracy improves when generating intermediate reasonin...
research
02/24/2022

Is Neuro-Symbolic AI Meeting its Promise in Natural Language Processing? A Structured Review

Advocates for Neuro-Symbolic AI (NeSy) assert that combining deep learni...
research
05/31/2022

GlanceNets: Interpretabile, Leak-proof Concept-based Models

There is growing interest in concept-based models (CBMs) that combine hi...
research
11/26/2019

Logical Interpretations of Autoencoders

The unification of low-level perception and high-level reasoning is a lo...
research
08/12/2021

Semantics-Native Communication with Contextual Reasoning

Spurred by a huge interest in the post-Shannon communication, it has rec...

Please sign up or login with your details

Forgot password? Click here to reset