You are tasked with training a classification model using Databricks AutoML. The dataset you are working with has several columns, but some of them are irrelevant to the classification task. Which parameter would you use to exclude specific columns during AutoML calculations?