Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factors

TL;DR

This paper explores a causal modeling approach to long-term fairness, introducing Tier Balancing to account for unobserved factors influencing future data distributions, and analyzes the feasibility of achieving fairness through interventions.

Contribution

It proposes a new fairness notion called Tier Balancing that considers unobserved causal factors and provides theoretical results on the limitations of interventions for long-term fairness.

Findings

One-step interventions are generally insufficient for long-term fairness.

Tier Balancing captures unobserved causal factors affecting future data.

The paper presents conditions for the possibility and impossibility of achieving long-term fairness.

Abstract

The pursuit of long-term fairness involves the interplay between decision-making and the underlying data generating process. In this paper, through causal modeling with a directed acyclic graph (DAG) on the decision-distribution interplay, we investigate the possibility of achieving long-term fairness from a dynamic perspective. We propose Tier Balancing, a technically more challenging but more natural notion to achieve in the context of long-term, dynamic fairness analysis. Different from previous fairness notions that are defined purely on observed variables, our notion goes one step further, capturing behind-the-scenes situation changes on the unobserved latent causal factors that directly carry out the influence from the current decision to the future data distribution. Under the specified dynamics, we prove that in general one cannot achieve the long-term fairness goal only through one-step interventions. Furthermore, in the effort of approaching long-term fairness, we consider the mission of "getting closer to" the long-term fairness goal and present possibility and impossibility results accordingly.