Beyond robustness: Resilience verification of tree-based classifiers

Calzavara, Stefano; Lorenzo, Cazzaro,; Lucchese, Claudio; Marcuzzi, Federico; Orlando, Salvatore

doi:10.1016/j.cose.2022.102843

Cited by 6 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Calzavara et al ( 2020a ) and Ranzato and Zanella ( 2020 ) take inspiration from the software verification field and use abstract interpretation , commonly used for static program analysis, for formal verification of tree ensembles. Calzavara et al ( 2022 ) propose a solution that verifies resilience , a generalization over robustness which considers all possible test sets that could be sampled.…”

Section: Decision Trees For Responsible Aimentioning

confidence: 99%

“…Other have used satisfiability modulo theory (SMT): the approaches byEinziger et al (2019),Sato et al (2020), andDevos et al (2021a) are similar and differ only in focus and implementation details Calzavara et al (2020a). and Ranzato and Zanella (2020) take inspiration from the software verification field and use abstract interpretation, commonly used for static program analysis, for formal verification of tree ensembles Calzavara et al (2022).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Decision trees: from efficient prediction to responsible AI

Blockeel¹,

Devos²,

Frénay‬³

et al. 2023

Front. Artif. Intell.

View full text Add to dashboard Cite

This article provides a birds-eye view on the role of decision trees in machine learning and data science over roughly four decades. It sketches the evolution of decision tree research over the years, describes the broader context in which the research is situated, and summarizes strengths and weaknesses of decision trees in this context. The main goal of the article is to clarify the broad relevance to machine learning and artificial intelligence, both practical and theoretical, that decision trees still have today.

show abstract

Section: Decision Trees For Responsible Aimentioning

confidence: 99%