How do you handle outliers in machine learning datasets?

Christopher

I want to understand how outliers affect machine learning models and predictions. What are the common methods used to detect and handle extreme or unusual data points? Can someone also explain when outliers should be removed versus retained?