---
title: Supremum Integral Probability Metric
url: https://www.emergentmind.com/topics/supremum-integral-probability-metric-supipm
type: topic
---

# Supremum Integral Probability Metric

The supremum Integral Probability Metric (supIPM) is a distributional fairness measure and theoretical tool formalizing the worst-case discrepancy between induced distributions of an algorithm—often with respect to subgroups distinguished by combinations of sensitive attributes—using an IPM evaluated over a class of discriminators. The supIPM is defined as the largest (supremum) value taken by an IPM between distributions corresponding to subgroups or subgroup-unions (“subgroup-subsets”), providing a principled, distribution-based approach to measuring and enforcing subgroup fairness under extensive intersectionality.

## 1. Formulation of supIPM in Subgroup Fairness

The supIPM arises as the following fairness divergence:

\[
\Delta_{𝒲,𝒢}(f) = \sup_{W ∈ 𝒲} \sup_{g ∈ 𝒢} | \mathbb{E}_{𝔓_{f,W}}[g] - \mathbb{E}_{𝔓_{f,W^{c}}}[g] |,
\]

where:
- $𝒲$ is a family of subgroup-subsets (i.e., sets composed from unions of intersectional subgroups determined by multiple sensitive attribute values),
- $𝔓_{f,W}$ is the conditional distribution of predictor outputs $f(X)$ given sensitive attributes in $W$,
- $𝔓_{f,W^{c}}$ is the analogous distribution for the complement,
- $𝒢$ is a class of discriminator functions (e.g., neural networks, Lipschitz, or parametric sigmoid-based functions),
- The inner IPM assesses the distributional distance between each subgroup $W$ and its complement, and the outer supremum takes the worst-case such gap across all elements of $𝒲$ [2510.21091].

This definition generalizes previous mean-based fairness notions to the distributional level and captures marginal, intersectional, and richer subgroup fairness regimes depending on the choice of $𝒲$.

## 2. Computational Scalability and the DRAF Algorithm

The direct implementation of supIPM can be computationally prohibitive due to the exponential growth of subgroup-subsets as the number of sensitive attributes increases. Each evaluation of supIPM requires computing an IPM for each $W ∈ 𝒲$, with $|𝒲|$ potentially exceeding $2^{q}$—the number of possible attribute combinations—making naive computation infeasible in practice for large $q$ or data sparsity in subgroups.

To address this, the Doubly Regressing Adversarial learning for Fairness (DRAF) algorithm introduces a surrogate fairness gap that is an explicit upper bound on supIPM, but can be optimized using a single adversary and weight vector:

- For each $i$, define a membership vector $c_i ∈ \{-1,1\}^M$ indicating membership in each subgroup-subset $W_m$.
- Introduce $v∈𝒮^M$, the $M$-dimensional unit sphere, to parameterize combinations of subgroup dependencies.
- Define the "doubly regressing" R² statistic:

\[
DR^2(f,v,g) = 1 - \frac{\sum_i (v^{\top} c_i - g(f_i))^2 - \sum_i (g(f_i) - \mu_v)^2}{\sum_i (v^{\top} c_i - \mu_v)^2}
\]

with $\mu_v = (1/n) \sum_i v^{\top} c_i$. Apply a Fisher z-transformation for numerical stability:

\[
DR_{n,𝒲,𝒢}(f) = \sup_{g ∈ 𝒢, v ∈ 𝒮^M} z-DR^2(f,v,g),
\]
\[
z-DR^2 = \log\left(\frac{1 + |DR^2|/2}{1 - |DR^2|/2}\right)
\]

This surrogate fairness gap is provably an upper bound on supIPM [2510.21091, Theorem 2], enabling adversarial optimization without explicit enumeration over all subgroup-subsets.

## 3. Theoretical Guarantees and Properties

DRAF’s surrogate fairness gap has key theoretical properties:
- It upper-bounds the original supIPM fairness gap $\Delta_{𝒲,𝒢}(f)$, ensuring that reducing the surrogate also reduces the true worst-case distributional gap.
- The approach allows optimization via a single adversary $g$ (from $𝒢$) and a single vector $v$ over the simplex, regardless of $|𝒲|$.
- Theoretical results establish that for a chosen $𝒢$, the difference in predictions across any $W ∈ 𝒲$ is captured via a modified R² regression fit, reducing the computational burden [2510.21091].

If the discriminator class $𝒢$ is rich enough, supIPM characterizes distributional fairness precisely; with simpler $𝒢$, it provides guaranteed control over particular classes of statistical disparities.

## 4. Empirical Methodology and Fairness-Accuracy Trade-offs

DRAF alternates minimization over predictor parameters and maximization over adversarial parameters $(g, v)$, optimizing the objective:

\[
\textrm{Loss} = \frac{1}{n} \sum_i \ell(y_i, f(x_i, s_i)) + \lambda \cdot DR_{n,𝒲,𝒢}(f)
\]

where $\lambda$ is a Lagrange multiplier trading off predictive accuracy and fairness.

Empirical results demonstrate:
- DRAF outperforms baseline methods (such as marginal fairness constraints or group-wise regularization) on benchmark datasets when the number of sensitive attributes $q$ is large and many intersectional subgroups are poorly represented.
- Trade-off assessments indicate that DRAF achieves favorable fairness (as measured by subgroup parity, marginal parity, and distributional metrics) without significantly compromising accuracy.
- Ablation studies show the importance of including all relevant subgroup-subsets in $𝒲$; limiting constraints to marginal fairness alone may leave fairness gaps unaddressed in small, intersectional subgroups [2510.21091].

## 5. Relation to General IPMs and Other Fairness Notions

SupIPM subsumes and generalizes earlier distributional and mean fairness measures:

- For $𝒢$ as the class of constant functions, supIPM reduces to worst-case mean parity disparity.
- For $𝒢$ as all $k$-Lipschitz functions, the IPM becomes the Wasserstein-$k$ distance; for RKHS or sigmoid function families, other fairness divergences arise.
- SupIPM’s distribution-based formalism supports both marginal and intersectional fairness and can handle subgroups with a wide range of sample sizes.
- Its design ensures theoretical consistency across the discrete-mean and full-distribution fairness landscape.

In the limit, as the discriminator class increases, the IPM becomes a strong differentiator for empirical distributions, making supIPM an effective tool for analyzing the extremal fairness achievable under adversarial training [2510.21091].

## 6. Limitations and Applicability

While DRAF and the underlying supIPM address computational issues inherent in intersectional subgroup fairness:
- The quality of approximation (tightness of the surrogate bound) depends on the richness of both the group set $𝒲$ and discriminator class $𝒢$; too narrow a choice may leave fairness violations undetected in some subgroups.
- Datasets with very small or empty subgroups require careful selection of $𝒲$ to ensure empirical tractability and statistical reliability.
- Interpretation of supIPM values relies on understanding the measure’s sensitivity to both $𝒲$ and $𝒢$ [2510.21091].

Nonetheless, the approach is demonstrably scalable and robust across a variety of real-world fairness tasks.

## 7. Broader Impact and Future Directions

The supIPM framework offers a principled mechanism for enforcing and measuring subgroup and intersectional fairness with rigorous distributional guarantees. DRAF and similar algorithms provide a computational toolkit for achieving these guarantees in modern, high-dimensional, and intersectional fairness scenarios. Further research may refine the selection and approximation of extremely large $𝒲$, as well as dynamically learn $𝒢$ tailored to data context. Extending supIPM-based certification to other types of statistical parity and causal fairness criteria remains an active direction.

**Summary Table: supIPM for Subgroup Fairness**

| Component        | Description                                            | Reference Section                    |
| ---------------- | ----------------------------------------------------- | ------------------------------------ |
| Mathematical Definition | $\Delta_{𝒲,𝒢}(f) = \sup_{W ∈ 𝒲} \sup_{g ∈ 𝒢} |𝔼_{𝔓_{f,W}}[g] - 𝔼_{𝔓_{f,W^c}}[g]|$ | 1     |
| Computational Surrogate | DRAF’s doubly regressing R² statistic with Fisher z-transformation as upper bound on supIPM | 2, 3 |
| Fairness Guarantee | Surrogate fairness gap provably upper-bounds true supIPM gap | 3     |
| Scalability         | Single adversary and vector optimization for arbitrarily large $𝒲$ | 2     |
| Empirical Efficacy  | Superior subgroup and marginal fairness under high intersectionality | 4     |
| Applicability       | Interpolation between mean, marginal, and intersectional fairness regimes | 5     |

The supIPM thus anchors a rigorous, distributional perspective on algorithmic fairness in settings with high-dimensional, intersectional sensitive attributes, enabling scalable, theoretically justified, and empirically robust learning algorithms [2510.21091].

Source: https://www.emergentmind.com/topics/supremum-integral-probability-metric-supipm