Principles and Practice of Explainable Machine Learning

Artificial intelligence (AI) provides many opportunities to improve private and public life. Discovering patterns and structures in large troves of data in an automated manner is a core component of data science, and currently drives applications in diverse areas such as computational biology, law and finance. However, such a highly positive impact is coupled with significant challenges: how do we

[lanius]

"Map of Explainability Approaches".

[misshiki]

説明可能なAI・機械学習の論文。前半は分野とツールの調査報告。後半では、適切な質問でモデルを説明する方法について考察”とのこと。ざっくりとして読んでいないが、必要なときに読み直したい。