Measuring Trustworthiness in Neuro-Symbolic Integration

by Andrea Agiollo and Andrea Omicini

Neuro-symbolic integration of symbolic and subsymbolic techniques represents a fast-growing AI trend aimed at mitigating the issues of neural networks in terms of decision processes, reasoning, and interpretability. Several state-of-the-art neuro-symbolic approaches aim at improving performance, most of them focusing on proving their effectiveness in terms of raw predictive performance and/or reasoning capabilities. Meanwhile, few efforts have been devoted to increasing model trustworthiness, interpretability, and efficiency – mostly due to the complexity of measuring effectively improvements in terms of trustworthiness and interpretability. This is why here we analyse and discuss the need for ad-hoc trustworthiness metrics for neuro-symbolic techniques. We focus on two popular paradigms mixing subsymbolic computation and symbolic knowledge, namely: (i) symbolic knowledge extraction (SKE), aimed at mapping subsymbolic models into human-interpretable knowledge bases; and (ii) symbolic knowledge injection (SKI), aimed at forcing subsymbolic models to adhere to a given symbolic knowledge. We first emphasise the need for assessing neuro-symbolic approaches from a trustworthiness perspective, highlighting the research challenges linked with this evaluation and the need for ad-hoc trust definitions. Then we summarise recent developments in SKE and SKI metrics focusing specifically on several trustworthiness pillars such as interpretability, efficiency, and robustness of neuro-symbolic methods. Finally, we highlight open research opportunities towards reliable and flexible trustworthiness metrics for neuro-symbolic integration.

• DOI: https://doi.org/10.15439/2023F6019
• URL: https://annals-csis.org/Volume_35/drp/6019.html
• URL open access: https://annals-csis.org/Volume_35/drp/pdf/6019.pdf

@inproceedings{keynote-fedcsis2023,
    author = {Agiollo, Andrea and Omicini, Andrea},
    booktitle = {Proceedings of the 18th Conference on Computer Science and Intelligence Systems},
    dblp = {conf/fedcsis/AgiolloO23},
    doi = {10.15439/2023F6019},
    editor = {Maria Ganzha and Leszek Maciaszek and Marcin Paprzycki and Dominik Ślęzak},
    eisbn = {978-83-967447-8-4},
    ieee = {10305942},
    iris = {11585/951750},
    isbn = {978-83-969601-0-8},
    isbn13 = {978-83-967447-9-1},
    issn = {2300-5963},
    month = sep,
    numpages = 10,
    pages = {1--10},
    scholar = {4358274351024166022},
    scopus = {2-s2.0-85179180821},
    series = {Annals of Computer Sciences and Information Systems},
    title = {Measuring Trustworthiness in Neuro-Symbolic Integration},
    url = {https://annals-csis.org/Volume_35/drp/6019.html},
    urlopenaccess = {https://annals-csis.org/Volume_35/drp/pdf/6019.pdf},
    urlpdf = {https://annals-csis.org/Volume_35/drp/pdf/6019.pdf},
    volume = 35,
    year = 2023
}