Fairness in Forecasting of Observations of Linear Dynamical Systems

Abstract

In machine learning, training data often capture the behaviour of multiple subgroups of some underlying human population. This behaviour can often be modelled as observations of an unknown dynamical system with an unobserved state. When the training data for the subgroups are not controlled carefully, however, under-representation bias arises. To counter under-representation bias, we introduce two natural notions of fairness in time-series forecasting problems: subgroup fairness and instantaneous fairness. These notions extend predictive parity to the learning of dynamical systems. We also show globally convergent methods for the fairness-constrained learning problems using hierarchies of convexifications of non-commutative polynomial optimisation problems. We also show that by exploiting sparsity in the convexifications, we can reduce the run time of our methods considerably. Our empirical results on a biased data set motivated by insurance applications and the well-known COMPAS data set demonstrate the efficacy of our methods.

Journal version of Zhou et al. [arXiv:2006.07315, AAAI 2021]

References in corpus (16)

• Equality of Opportunity in Supervised Learning
• Inherent Trade-Offs in the Fair Determination of Risk Scores
• Bias and Discrimination in AI: a cross-disciplinary perspective
• Avoiding Discrimination through Causal Reasoning
• On the relation between accuracy and fairness in binary classification
• Learning Linear Dynamical Systems with Semi-Parametric Least Squares
• Assessing Fairness in the Presence of Missing Data
• Empirical observation of negligible fairness-accuracy trade-offs in machine learning for public policy
• Dynamic fairness - Breaking vicious cycles in automatic decision making
• Causal Conceptions of Fairness and their Consequences
• Representation Matters: Assessing the Importance of Subgroup Allocations in Training Data
• Does enforcing fairness mitigate biases caused by subpopulation shift?
• Fairness without Imputation: A Decision Tree Approach for Fair Prediction with Missing Values
• Exploiting term sparsity in Noncommutative Polynomial Optimization
• Online Learning of the Kalman Filter with Logarithmic Regret
• Protecting Consumers Against Personalized Pricing: A Stopping Time Approach