Geometric Origins of Bias in Deep Neural Networks: A Human Visual System Perspective
This addresses bias in AI systems for fairness and reliability, offering a theoretical foundation for more equitable models.
The paper tackles the problem of bias formation in deep neural networks by analyzing how differences in geometric complexity of class-specific perceptual manifolds lead to varying recognition capabilities across categories, providing a novel geometric perspective on bias.
Bias formation in deep neural networks (DNNs) remains a critical yet poorly understood challenge, influencing both fairness and reliability in artificial intelligence systems. Inspired by the human visual system, which decouples object manifolds through hierarchical processing to achieve object recognition, we propose a geometric analysis framework linking the geometric complexity of class-specific perceptual manifolds in DNNs to model bias. Our findings reveal that differences in geometric complexity can lead to varying recognition capabilities across categories, introducing biases. To support this analysis, we present the Perceptual-Manifold-Geometry library, designed for calculating the geometric properties of perceptual manifolds. The toolkit has been downloaded and installed over 4,500 times. This work provides a novel geometric perspective on bias formation in modern learning systems and lays a theoretical foundation for developing more equitable and robust artificial intelligence.