Energy shields for fairness

conference paper Energy shields for fairness published yes Filip Cano Cordoba author 708cad98-e86a-11ef-8098-bdae2d7c6af10000-0002-0783-904X Thomas A Henzinger author 40876CD8-F248-11E8-B48F-1D18A9856A870000-0002-2985-7724 Konstantin Kueffner author 8121a2d0-dc85-11ea-9058-af578f3b45150000-0001-8974-2542 ToHe department FAccT: Conference on Fairness, Accountability and Transparency Vigilant Algorithmic Monitoring of Software project Runtime fairness is not a one-time constraint but a dynamic property evaluated over a sequence of decisions. To ensure fairness at runtime, it is necessary to account for past decisions, information neglected by conventional, static classifiers. Traditional fairness shields enforce runtime fairness abruptly, by intervening deterministically whenever a sequence of decisions violates the target for a running fairness measure. This motivates our main conceptual contribution: energy shields. An energy shield is a novel, lightweight, adaptive controller that monitors a sequence of decisions and intervenes probabilistically to ensure runtime fairness smoothly, by utilizing physics-inspired energy functions to nudge the sequence toward fairness: the more unfair the decisions, the stronger the nudging force becomes. This makes energy shields the first fairness shields to provide both short-term safety and long-term liveness guarantees. Safety ensures that the running fairness measure stays within a running target interval with high probability, and liveness ensures that the limit of the fairness measure lies within the limit target interval. Intuitively, the short-term specifies the tolerated fairness values and the long-term specifies the desired fairness values. We also provide a synthesis procedure for constructing the least intrusive energy shield for a given target specification, and demonstrate its efficiency experimentally. We evaluate our energy shields against existing fairness shields through the lens of short- and long-term fairness. https://research-explorer.ista.ac.at/download/22321/22348/2026_ACMFACCT_Cano.pdf application/pdfno Association for Computing Machinery2026Montreal, Canada eng Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency 2605.2492610.1145/3805689.3806807 4243 - 4275 Cano Cordoba, Filip, et al. “Energy Shields for Fairness.” Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency, Association for Computing Machinery, 2026, pp. 4243–75, doi:<a href="https://doi.org/10.1145/3805689.3806807">10.1145/3805689.3806807</a>. Cano Cordoba F, Henzinger TA, Kueffner K. Energy shields for fairness. In: Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency. Association for Computing Machinery; 2026:4243-4275. doi:<a href="https://doi.org/10.1145/3805689.3806807">10.1145/3805689.3806807</a> F. Cano Cordoba, T.A. Henzinger, K. Kueffner, in:, Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency, Association for Computing Machinery, 2026, pp. 4243–4275. Cano Cordoba, F., Henzinger, T. A., & Kueffner, K. (2026). Energy shields for fairness. In Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency (pp. 4243–4275). Montreal, Canada: Association for Computing Machinery. <a href="https://doi.org/10.1145/3805689.3806807">https://doi.org/10.1145/3805689.3806807</a> Cano Cordoba F, Henzinger TA, Kueffner K. 2026. Energy shields for fairness. Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency. FAccT: Conference on Fairness, Accountability and Transparency, 4243–4275. F. Cano Cordoba, T. A. Henzinger, and K. Kueffner, “Energy shields for fairness,” in Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency, Montreal, Canada, 2026, pp. 4243–4275. Cano Cordoba, Filip, Thomas A Henzinger, and Konstantin Kueffner. “Energy Shields for Fairness.” In Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency, 4243–75. Association for Computing Machinery, 2026. <a href="https://doi.org/10.1145/3805689.3806807">https://doi.org/10.1145/3805689.3806807</a>. 223212026-07-14T05:32:45Z2026-07-16T09:24:01Z