Is Machine Learning Unsafe and Irresponsible in Social Sciences? Paradoxes and Reconsidering from Recidivism Prediction Tasks

Abstract

The paper addresses some fundamental and hotly debated issues for high-stakes event predictions underpinning the computational approach to social sciences. We question several prevalent views against machine learning and outline a new paradigm that highlights the promises and promotes the infusion of computational methods and conventional social science approaches.

References in corpus (3)

• Underspecification Presents Challenges for Credibility in Modern Machine Learning
• Stop ordering machine learning algorithms by their explainability! A user-centered investigation of performance and explainability
• Towards Quantification of Explainability in Explainable Artificial Intelligence Methods