From the course: Machine Learning with SageMaker by Pearson

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Using SageMaker Clarify for bias detection and interpretability

Using SageMaker Clarify for bias detection and interpretability - Amazon SageMaker Tutorial

From the course: Machine Learning with SageMaker by Pearson

Start my 1-month free trial Buy for my team

Using SageMaker Clarify for bias detection and interpretability

“

Bias in the machine learning world is when a model prefers one particular class over another. There's a tool within SageMaker called Clarify that allows you to automate the detection of bias in data and models. Clarify allows you to explain the predictions that are coming allows you to explain the predictions from a machine learning model. that are coming from a machine learning model. And it is integrated with SageMaker workflows so that you can, for example, put it into a pipeline and determine why a particular prediction is being made. So let's talk about this. Why is it important to detect bias? It ensures fairness across data sets as well as models. It ensures that there is an equitable distribution across demographic groups. And this is demographic here, it could be a sort of a logical construct or an actual human demographic group, as we'll see up here in a moment in the example use case. This can mitigate legal as well as ethical risks associated with biased decisions. So if…

Contents