Europejski Urząd Nadzoru Bankowego, Uczenie maszynowe dla modeli opartych na ratingach wewnętrznych

EBA/REP/2023/28

AUGUST 2023

MACHINE LEARNING FOR IRB MODELS

FOLLOW-UP REPORT FROM THE CONSULTATION ON THE DISCUSSION PAPER ON MACHINE LEARNING FOR IRB MODELS

1. Executive Summary

The aim of this follow-up report is to summarise the main conclusions from the consultation on the discussion paper (DP) on Machine learning (ML) used in the context of internal ratings-based (IRB) models. In addition, this report discusses the interaction between prudential requirements on IRB models and two other legal frameworks that have an impact on internal credit risk models that use ML techniques, namely the General Data Protection Regulation (GDPR) and the Artificial Intelligence (AI) Act.

The exponential increase in data availability and storing capacity coupled with the improvements in computing power of recent years provide an opportunity to use ML models. As presented in the DP, the follow-up report focuses on the more complex models than traditional techniques, such as regression analysis or simple...