Explanations when optimized towards prediction.

<div><p>This study introduces a post-hoc approach for explainable machine learning on graph-structured data by identifying relevant subgraphs and their corresponding subgraph patterns, the graphlet motifs. Unlike traditional motif detection methods, which rely solely on statistical occur...

وصف كامل

محفوظ في:

التفاصيل البيبلوغرافية
المؤلف الرئيسي:	Bettina Soós (10860933) (author)
مؤلفون آخرون:	Gonzalo Nápoles (22127303) (author), Pieter Spronck (13717717) (author), Çiçek Güven (8541117) (author)
منشور في:	2025
الموضوعات:	Biochemistry Neuroscience Evolutionary Biology Space Science Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified statistical occurrence frequencies model &# 8217 core reasoning behind overly complex subgraphs subgraph coverings correspond objective optimization problem corresponding subgraph patterns identifying relevant subgraphs method outperforms state hence avoiding redundant graphs </ p deemed key properties also contribute meaningfully explainable machine learning possible subgraph space determining graphlet explanations graph prediction tasks machine learning subgraph explanations relevant due hard problem graphlet motifs covering subgraph explanatory subgraphs learning task explanatory subgraph xlink "> study introduces structured data still capturing single motifs resulting motifs rely solely prediction task possible motifs motifs found hoc approach genetic algorithm generating explanations desirable explanations computationally difficult comparing performance building blocks art techniques art methodologies also necessary
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

Explanations when optimized towards prediction.

مواد مشابهة