Algorithmic justice and its limitations: An impossibility theorem

Algorithmic justice in learning models refers to the application of ethical and fairness principles in the development and use of machine learning algorithms. These algorithms are used to make predictions or decisions based on data, and can have a significant impact on people's lives, for example, in the field of job candidate selection, loan approval, assessment of the risk of committing crimes, among others. Algorithmic justice seeks to avoid discrimination and bias in data-driven decision making. However, there are multiple definitions and perspectives on what algorithmic justice means and how it can be achieved. These definitions can be so heterogeneous that it may be impossible to obtain two or more of these simultaneously in a single estimate. In this blog, based on the article On Fairness and Calibration [1] two different definitions of algorithmic fairness are explained and a graphical demonstration of the impossibility of achieving both is given.

Concepts of fairness

Two definitions of fairness are considered: Equalized Odds and Calibration. For this, we define a classifier h that takes values between 0 and 1, a protected variable A with two possible values 1 and 2 (e.g. female or male sex), and as the actual or observed value of the classification. The classifiers are defined as the constraints of the initial classifier with respect to the two possible values of A. In addition, the values of generalized false positive rate and  generalized true positive rate are defined as:

With this it is possible to define Equalized Odds: a classifier satisfies Equalized Odds if.

On the other hand, it is possible to state that a classifier is perfectly calibrated if for every value . Moreover, it satisfies the concept of fairness by calibration if the classifier's restrictions on the categories of the protected variable are also perfectly calibrated. That is,

Proof of impossibility 

In the articles On Fairness and Calibration [1] y Inherent trade-offs in the fair determination of risk scores [2] the impossibility of having both definitions of algorithmic fairness on classifiers . For this, the geometric implications of both concepts of fairness are initially observed.

Given a classifier h, it is possible to represent the performance of the model in a plane as a tuple. If one has two classifiers with the same value for the generalized false positive rate, it means that the points in this space must lie on the same vertical line as shown in Figure 1. On the other hand, if  they have the same value for the generalized true positive rate, then the points must lie on the same horizontal line, as shown in Figure 2.

Figure 1

Figure 2

Therefore, to have Equalized Odds both points obtained must coincide, that is to say that

Now, a perfectly calibrated classifier can be represented by a calibration curve as shown in Figure 3. In this curve, the observations X are separated into different sets or bins depending on the probability assigned to the observation with the classifier h. For example, if we consider all the observations assigned a probability of 0.1, we have that

   Figure 3

By the definition of a perfectly calibrated classifier we have that for any value p ∈ [0,1]:

Where |C| represents the cardinality of the set C.

If this result is used on all the values taken from p 

From this, it is possible to obtain the following equality

Taking , one has that for a perfectly calibrated classifier,

That is to say that there is a linear relationship between the values of and . If we now have two perfectly calibrated classifiers with respective values of , the points obtained in the plane must lie on the respective straight lines defined by the equation obtained. Figure 4 shows the straight lines on which the tuples must be found for respectively.

Figure 4

If, in addition to requesting the calibration of both classifiers, we want them to satisfy Equalized Odds, the points must coincide, so they must be at a point where the two straight lines intersect. If then the only point that satisfies both conditions is the point of intersection of both straight lines, which is the tuple (0,1), which represents the perfect classifier. This concludes the proof of impossibility.

References

[1] G. Pleiss, M. Raghavan, F. Wu, J. Kleinberg, K. Q. Weinberger. On Fairness and Calibration. arXiv:1709.02012, 2017.

[2] J. Kleinberg, S. Mullainathan, and M. Raghavan. Inherent trade-offs in the fair determination of risk scores. In Innovations in Theoretical Computer Science. ACM, 2017.

Tags
Supervised learning Calibration Algorithmic Justice

Newsletter

Get information about Data Science, Artificial Intelligence, Machine Learning and more.

Recent articles

In the Blog articles, you will find the latest news, publications, studies and articles of current interest.

Algorithmic Justice

Justicia en los Modelos de Inteligencia Artificial: Nueva Perspectiva Basada en el Re-diseño de Algoritmos

En los últimos años, los modelos de inteligencia artificial han demostrado un potencial increíble para transformar industrias, desde la salud hasta las finanzas. Sin embargo, también han expuesto un problema preocupante: el sesgo algorítmico.

Machine Learning

Inferencia Robusta y Cuantificación de Incertidumbre para la Toma de Decisiones Basada en Datos

Los modelos de aprendizaje automático se han convertido en herramientas esenciales para la toma de decisiones en sectores críticos como la salud, las políticas públicas y las finanzas. Sin embargo, su aplicación práctica enfrenta dos grandes desafíos: el sesgo de selección en los datos y la cuantificación adecuada de la incertidumbre.

Redes Neuronales

El Potencial Impacto del Aprendizaje de Máquinas en el Diseño de las Políticas Públicas en Colombia: Una década de experiencias

Este blog es un resumen extendido del articulo Riascos, A. (2025).1 Desde el inicio de la llamada tercera ola de redes neuronales (Goodfellow et al., (2016)), en la primera década de este siglo, se ha generado una gran esperanza en las posibilidades de la inteligencia artificial para transformar todas las actividades humanas. Asimismo, se han levantado alertas sobre los riesgos que conlleva la introducción de esta nueva tecnología (Bengio et al., (2024)).

Deep Learning

Explorando Redes Neuronales en Grafos para la Clasificación de Asentamientos Informales en Bogotá, Colombia

Los asentamientos informales son definidos como áreas residenciales cuyos habitantes no poseen tenencia legal de las tierras, los barrios carecen de servicios básicos e infraestructura urbana y no cumplen con requisitos de planificación, así como se pueden encontrar en zonas de peligro ambiental y geográfico (ONU, 2015).

Technology

Reinforcement Learning para Optimización de Portafolios

En el contexto de los mercados financieros, la optimización de portafolios consiste en identificar la combinación óptima de activos para maximizar la relación retorno-riesgo. No obstante, esta toma de decisiones se realiza en un entorno de incertidumbre, ya que el comportamiento de los activos no es estacionario a lo largo del tiempo.

Technology

Clustering de datos genómicos

La secuenciación de RNA es una técnica que permite analizar la actividad de los genes en una muestra, como sangre, cerebro u otro tejido animal. Actualmente, es una de las herramientas más utilizadas en biología computacional y medicina, ya que facilita el estudio del impacto de las enfermedades en la expresión génica, lo que, a su vez, afecta la síntesis de proteínas y, en consecuencia, el funcionamiento celular.