Course 1
The concept of bias
The concept of bias
Deepen your knowledge of the basics with practical exercises to increase your learning success. Please note that the exercises are based on specific sections of the Basics page of this unit.
This course builds on the “Different Types of Bias” section of the basic page.
Use this course to deepen your understanding of bias.
In the first assignment we look at unconscious bias in hiring decisions. In the second assignment, we will deepen the different types of bias in the ML lifecycle that we have learned about. To do this, we will give you different scenarios to which you can then assign a particular type of bias.
Task 1.1
Unconscious bias -
Do you know the facts?
Task 1.1
Unconscious biases influence your actions, whether you are hiring or lending. For example, each of us has an often unconscious tendency to associate certain terms with one gender or the other. Many people spontaneously associate male first names with the word ‘career’ and female first names with the word ‘family’. You can try this test yourself: The Harvard Implicit Association Test.
Our first task is to see if you can identify unconscious bias behind hiring decisions. Unconscious bias in the workplace is when people make judgements that are at least partly influenced by gender, ethnicity, race or other factors without being aware of it. This is usually based on social stereotypes or personal experience.
Take this quiz to test your knowledge of unconscious bias.
Rather than relying on human assumptions and decisions, which are often made with unconscious bias, AI systems in machine learning refer to models generated from a variety of data sources and parameters. Many of those who develop and deploy AI systems make a blanket assumption about the objectivity and neutrality of algorithmic decisions. However, they have already learned that this is not always the case.
Source of the quiz questions: everywoman
Task 1.2
Different types of bias
Task 1.2
As you know, bias can occur at any stage of the ML lifecycle. We have defined eleven specific types of bias in the ML lifecycle on the basics page. The following exercise will help you illustrate the definitions of each type of bias.
In the following quiz you will be given examples of certain types of bias. Your task is to match each example with its type of bias.
Awesome! You have completed the first course!