Revealing Neural Network Bias to Non-Experts Through Interactive Counterfactual Examples
This addresses the issue of algorithmic bias for non-experts, though it is incremental as it builds on existing methods like counterfactual examples.
The paper tackled the problem of non-experts lacking control in uncovering social biases like gender bias in neural networks, and presented CEB, an interactive visualization tool that combines counterfactual examples and abstraction to help detect such biases, with initial findings from an expert panel of six.
AI algorithms are not immune to biases. Traditionally, non-experts have little control in uncovering potential social bias (e.g., gender bias) in the algorithms that may impact their lives. We present a preliminary design for an interactive visualization tool CEB to reveal biases in a commonly used AI method, Neural Networks (NN). CEB combines counterfactual examples and abstraction of an NN decision process to empower non-experts to detect bias. This paper presents the design of CEB and initial findings of an expert panel (n=6) with AI, HCI, and Social science experts.