Model-Agnostic Interpretation via Feature Perturbation Visualization

  • Wilson E. Marcílio Júnior UNESP
  • Danilo Medeiros Eler UNESP
  • Fabrício Breve UNESP


As machine learning algorithms increasingly replace traditional approaches, ensuring their reliability becomes crucial in applications where incorrect decisions can lead to serious consequences. This work proposes a novel model-agnostic in-terpretation approach using feature perturbations, along with a validated visualization tool. The approach enables better understanding of model decisions by domain experts, facilitating effective decision-making in real-world applications.
Como Citar

Selecione um Formato
MARCÍLIO JÚNIOR, Wilson E.; ELER, Danilo Medeiros; BREVE, Fabrício. Model-Agnostic Interpretation via Feature Perturbation Visualization. In: CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), 36. , 2023, Rio Grande/RS. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2023 . p. 19-24.