Avoiding Structural Failure Modes in Tabular Fair SSL: Online Primal-Dual Allocation under Confidence Gating

TL;DR

This paper introduces OPDA, an online controller for fair semi-supervised learning on tabular data, which avoids failure modes and adapts fairness-utility trade-offs without dataset-specific tuning.

Contribution

The paper proposes OPDA, a novel online primal-dual controller that dynamically balances fairness and utility in tabular SSL, preventing failure modes caused by static weighting.

Findings

OPDA mitigates failure modes in tabular fair SSL.

OPDA achieves competitive fairness-utility trade-offs on benchmarks.

OPDA preserves utility while maintaining fairness across datasets.

Abstract

Semi-supervised learning (SSL) enables prediction with limited labels, but high-stakes tabular applications (medical, credit, recidivism) require statistical fairness guarantees. We identify a structural conflict in tabular fair SSL through a diagnostic stress test: under confidence-gated pseudo-labeling, moment-matching fairness regularizers can trigger two failure modes -- Masking Collapse (fairness erodes confidence, starving pseudo-labels) and Trivial Saturation (drift to constant predictors). We propose Online Primal-Dual Allocation (OPDA), an online controller that schedules fairness and entropy-based stability penalties using violation, risk, and pseudo-label health signals, avoiding per-dataset selection of a fixed fairness weight within this diagnostic regime. On the evaluated tabular benchmarks (Adult, ACSIncome, COMPAS), OPDA mitigates the degenerate regimes observed under static weighting and simple single-signal adaptive baselines. On Adult and COMPAS, it yields non-degenerate operating points competitive with the empirical static-$\lambda$ frontier; on ACSIncome, it preserves utility with a wider fairness-utility spread. Relative to OPDA-lite, the full controller mainly shifts the operating point toward higher utility on ACSIncome, while Adult highlights the fairness-utility trade-off between the two variants. These results position OPDA as a calibration-free controller for non-degenerate operating points in tabular fair SSL without per-dataset tuning.