Learning Informative Representation for Fairness-aware Multivariate Time-series Forecasting: A Group-based Perspective

TL;DR

This paper introduces FairFor, a novel fairness-aware multivariate time-series forecasting framework that generates group-independent and group-relevant representations to ensure equitable attention to all variables, improving fairness and accuracy.

Contribution

The paper proposes a new adversarial learning framework, FairFor, which infers variable groups and generates informative representations to promote fairness in multivariate time-series forecasting.

Findings

FairFor effectively improves fairness in forecasting across four datasets.

The framework enhances model performance while reducing variable bias.

Extensive experiments validate the superiority of FairFor over existing methods.

Abstract

Performance unfairness among variables widely exists in multivariate time series (MTS) forecasting models since such models may attend/bias to certain (advantaged) variables. Addressing this unfairness problem is important for equally attending to all variables and avoiding vulnerable model biases/risks. However, fair MTS forecasting is challenging and has been less studied in the literature. To bridge such significant gap, we formulate the fairness modeling problem as learning informative representations attending to both advantaged and disadvantaged variables. Accordingly, we propose a novel framework, named FairFor, for fairness-aware MTS forecasting. FairFor is based on adversarial learning to generate both group-independent and group-relevant representations for the downstream forecasting. The framework first leverages a spectral relaxation of the K-means objective to infer variable correlations and thus to group variables. Then, it utilizes a filtering&fusion component to filter the group-relevant information and generate group-independent representations via orthogonality regularization. The group-independent and group-relevant representations form highly informative representations, facilitating to sharing knowledge from advantaged variables to disadvantaged variables to guarantee fairness. Extensive experiments on four public datasets demonstrate the effectiveness of our proposed FairFor for fair forecasting and significant performance improvement.