Mitigating Bias in Graph Hyperdimensional Computing

TL;DR

This paper introduces FairGHDC, a fairness-aware training framework for graph hyperdimensional computing that reduces bias and maintains efficiency, outperforming existing methods in fairness metrics while being faster.

Contribution

We propose a novel fairness-aware training method for graph HDC that mitigates bias without altering the encoder or using backpropagation.

Findings

Significantly reduces demographic-parity and equal-opportunity gaps.

Maintains comparable accuracy to fairness-aware GNNs.

Achieves up to 10x faster training on GPU.

Abstract

Graph hyperdimensional computing (HDC) has emerged as a promising paradigm for cognitive tasks, emulating brain-like computation with high-dimensional vectors known as hypervectors. While HDC offers robustness and efficiency on graph-structured data, its fairness implications remain largely unexplored. In this paper, we study fairness in graph HDC, where biases in data representation and decision rules can lead to unequal treatment of different groups. We show how hypervector encoding and similarity-based classification can propagate or even amplify such biases, and we propose a fairness-aware training framework, FairGHDC, to mitigate them. FairGHDC introduces a bias correction term, derived from a gap-based demographic-parity regularizer, and converts it into a scalar fairness factor that scales the update of the class hypervector for the ground-truth label. This enables debiasing directly in the hypervector space without modifying the graph encoder or requiring backpropagation. Experimental results on six benchmark datasets demonstrate that FairGHDC substantially reduces demographic-parity and equal-opportunity gaps while maintaining accuracy comparable to standard GNNs and fairness-aware GNNs. At the same time, FairGHDC preserves the computational advantages of HDC, achieving up to about one order of magnitude ($\approx 10\times$) speedup in training time on GPU compared to GNN and fairness-aware GNN baselines.