Peer Review #2 of "A step toward building a unified framework for managing AI bias (v0.1)"
2023
Integrating Artificial Intelligence (AI) into every aspect of our experiences has significantly improved living standards.However, AI's efforts are being thwarted by concerns about the rise of biases and unfairness.Conditions become further exacerbated due to multiple interlinked forms of biases and strategies.The problem advocates strongly for the existence of an organized strategy for tackling potential biases.This paper thoroughly evaluates existing knowledge to enhance Bias Management, which will serve as a foundation for creating a unified framework to address any bias and its subsequent mitigation method throughout the AI development pipeline.We map the Software Development Life Cycle (SDLC), Machine Learning Life Cycle (MLLC) and Cross Industry Standard Process for Data Mining (CRISP-DM) process model to have a general understanding of how phases in these development processes are related to each other.The map should benefit researchers from multiple technical backgrounds.Biases are categorized into three distinct classes; Pre-existing, Technical and Emergent Bias, and subsequently, three mitigation strategies; Conceptual, Empirical and Technical, along with Fairness Management Approaches; Fairness Sampling, Learning and Certification.The recommended practices for debias and overcoming challenges encountered further set directions for successfully establishing a unified framework.