Ethical Bias check algorithms

Ethical Bias check algorithms

Description	Ethical Bias check algorithms
Version as of	8.4
Application	Pega Customer Decision Hub
Capability/Industry Area	Testing and Simulation

Ethical Bias check algorithms

This document describes the algorithms and metrics used in the ethical bias check. This document is aimed at the audience that wants to know exactly how the ethical bias check works.

Use case examples

The ethical bias check is a simulation type that can be used to test if unwanted bias is introduced in the Next-Best-Action strategy. This document describes the use of the rate ratio and the Gini coefficient in detecting bias.

Before you begin

• Get acquainted with the ethical bias check.
• Read how to ensure fairness and compliance with Pega’s Ethical Bias Check, on Pega documentation.
• Learn how to configure ethical bias checks, on Pega documentation.
• Study the ethical bias concept on Pega Academy.

Bias fields

The metric used to detect bias is Rate ratio for categorical or binary fields and is the Gini coefficient for numeric (ordinal) fields.

The following Pega data types are allowed as bias types, these fields can be added as bias fields in the Ethical Bias Policy. Please note that only users with pzBiasPolicyConfiguration privilege added to their access group have permission to configure the bias policy.

Pega field type	Bias treated as	Bias measure	Examples	Remarks
Text / TrueFalse / Integer	Categorical	Rate ratio	Gender, Ethnicity	Bias will be checked for each categorical value. If there are many categorical values, only the first 20 most frequent values are checked for bias and the remaining values are treated as a rest category. Missing values are ignored. Note: Integer is allowed as a Pega field type but it should only be used if the information contained in the field is not ordinal, otherwise select Numeric treatment.
Decimal / Double / Integer	Numeric	Gini coefficient	Age	Missing values are ignored.
Date / Datetime / Identifier / Password / TextEncypted / TimeOfDay	Not allowed as bias fields

Rate ratio or disparate impact

Rate ratio is the measure of discrimination used for binary or categorical fields and is based on a contingency table. It's also called risk ratio, for example in the medical field, but we use rate ratio as in our case being selected for an action is not a risk. Another related term for this is disparate impact, although there is a small difference with that definition. In this case, there is no defined direction for the bias, it could be a bias in both directions, but apart from that the definition is the same.

The rate ratio is calculated from a 2x2 contingency table.

An example of a contingency table is given below:

	female	male
selected for action	500	1,000
not selected for action	20,000	18,000

Rate ratio for female = [500  / (500+20,000) ] / [1,000 /(1,000+18,000] = 0.024 / 0.052 = 0.46

Rate ratio for male    = [1000 / (1000+18,000) ] / [500 /(20,000+500] = 0.66 / 0.47 = 2.16

The rate ratio ranges between 0 and infinity.

• A value of 1 indicates there is no bias at all
• Values smaller or greater than 1 indicate there is bias

Missing value treatment: cases containing missing values are ignored

References:

https://en.wikipedia.org/wiki/Odds_ratio

https://en.wikipedia.org/wiki/Contingency_table

https://en.wikipedia.org/wiki/Relative_risk

Setting a bias threshold for the rate ratio

The bias thresholds are used to set an allowed range for the rate ratio. If the rate ratio is outside of this allowed range with a confidence interval of 95%, then this will be signaled as significant bias. For each bias threshold setting, no difference is made between a positive shift (towards > 1) and a negative shift (towards < 1):

Bias threshold	Allowed rate ratio range
no bias allowed	any bias with a 95% confidence for the rate ratio to be greater than 1 or less than 1 will be detected
very light	0.90 - 1.11
light	0.80 - 1.25
heavy	0.66 - 1.50
very heavy	0.50 - 2.00
all bias allowed	no bias detection

How bias is detected with respect to the set threshold and confidence interval

GINI measure

Illustration of how the Gini coefficient is used to measure the difference between two distributions

The GINI measure is used to calculate the bias in an ordinal field, for example age. It expresses how much age can be used to discriminate between the group that receives action X and those that don't receive action X.

Missing value treatment: cases containing missing values are ignored

Range: the GINI coefficient ranges between 0 and 1 (or 0 and 100%)

Note that the GINI coefficient is directly related to the Area Under the ROC Curve (AUC) in the way that AUC = (GINI+1) / 2.

References:

https://en.wikipedia.org/wiki/Gini_coefficient

Setting a bias threshold for the Gini coefficient

The bias thresholds are used to set an allowed range for the Gini coefficient. If the measured Gini coefficient is outside of this allowed range with a confidence interval of 95%, then this will be signaled as significant bias.

Bias threshold	Allowed Gini coefficient range
no bias allowed	any bias with a 95% confidence for the Gini coefficient to be greater than 0 will be detected
very light	< 0.10
light	< 0.20
heavy	< 0.50
very heavy	< 0.70
all bias allowed	no bias detection

Methods for determining the confidence intervals

For the rate ratio, the confidence interval is calculated using the following approximation on the error on log(RR):

(for the 95% confidence interval z= 1.96).

The confidence interval for the Gini coefficient is calculated using Delong's method. This function is also used by well-known packages such as pROC. The error on the Gini coefficient is 2x the error on the AUC.

Dynamic System Setting for the confidence level used to determine the interval width

In Pega Platform, the confidence level for detecting bias above threshold can be set through a Dynamic System Setting (DSS). The default value is 0.9999 (or 99.99%). Note that lowering this level increases the probability of false alerts on bias detected.

Pega-DecisionEngine

decision/simulation/ethicalbias/confidenceinterval

0.9999

Ethical bias report

For each category: 4 contingency numbers, 2 rates, 1 rate ratio

For each numeric field: 1 Gini coefficient

The numbers that can be part of the bias report are illustrated by this example:

Bias exceeds threshold	Issue	Group	Action	Field	Category	Rate ratio	(Rate bootstrapped)	Rate ratio 95% confidence interval	Rate ratio allowed range	Receiving action, #category	Receiving action, #rest	Receiving action, category rate	Not receiving action, #category	Not receiving action, #rest	Not receiving action, category rate	#category total	Gini coefficient	(Gini coefficient bootstrapped)	Gini 95% confidence interval	Gini coefficient threshold	Receiving action, average value	Not receiving action, average value
Yes	Sales	Phones	iPhoneX	Gender	Female	0.38	0.37	0.37-0.39	0.80-1.25	1179	6662	15%	9592	15128	38%	10,771 (33%)
Yes					Male	2.60	2.55	2.53-2.57	0.80-1.25	6662	1179	38%	15128	9592	15%	21,790 (66%)
					(Missing)											0
No				Ethnicity	Ethnicity-White
No					Ethnicity-Asian
No					Ethnicity-Other
					(Missing)
Yes				Age													45%	44.5	44.4-44.6	40%	38.1	39.2
					(Missing)											1234
Yes	Risk	Loan	Personal Loan	Gender	Female				0.80-1.25
Yes					Male				0.80-1.25
					(Missing)
No				Ethnicity	Ethnicity-White
No					Ethnicity-Asian
No					Ethnicity-Other
Yes				Age

A business solution that automates workflows in your organization. Multiple, unrelated applications can be hosted in the same system. You create an application by running the New Application wizard.

An application-specific access control element that is associated with a class and an access role. Using privilege rules in an application is optional, but they can offer improved control over access than access roles alone.

A set of users who receive access to application or network resources through their membership in the group. When a user is added to an access group, that user receives the permissions that are associated with the group. If a user belongs to more than one access group, that user acquires the set of access permissions that are associated with each group. You can create multiple access groups within an application to convey different levels of access based on user roles.

Actual work that is performed by application users and automations that consists of a sequence of tasks and leads to an intended business goal. An example of a case is a process of hiring a particular job candidate, where an HR worker collects necessary documents, a hiring manager conducts a job interview, and an application sends an automated email with onboarding information after the candidate is approved. To create a case, you first build a template of your business process that is a case type, and then apply the template in multiple cases, for example, to review applications from many candidates.

The top level of the three-level organization hierarchy that is available for use in all of your applications. The organizational structure affects the management reports, statistics, and rules that are available to users in that organization. One system can support multiple organizations. An organization is an instance of the Data-Admin-Organization class.

A target range of values. When a threshold is exceeded, an alert is generated. You can manage threshold settings and specify whether alerts display by updating the prconfig.xml file, or by using dynamic system settings.

(Deprecated) A configuration of panels (areas) in a harness that support a composite portal. Existing panel sets continue to function. As a best practice, upgrade panel sets to screen layouts.

A Pega Platform method, distinct from a Java method. An operation that can be performed by one step of an activity ( Rule-Obj-Activity rule type), or by a Parse Structured rule ( Rule-Parse-Structured rule type).

An element that defines a Java function and makes the function callable from activities and expressions. A function is an instance of the Rule-Utility-Function rule type.

A data instance that is used to change the performance of Pega Platform. Dynamic system settings that have Pega-Engine as the owning ruleset and key names that start with prconfig are known as prconfig settings. For easier maintenance, create a dynamic system setting data instance and set the value of the data instance to a prconfig setting, instead of directly updating prconfig settings.