ML jobs- aggregations

Hey, Im unable to understand how the several aggregation techniques we use for ML jobs. Is there any blog which explains how these aggregation techniques work or can anyone explain about it?
thanks in advance

Docs: https://www.elastic.co/guide/en/elastic-stack-overview/current/ml-configuring-aggregation.html

Blog: https://www.elastic.co/blog/custom-elasticsearch-aggregations-for-machine-learning-jobs

Can you explain the difference between low_count and low_sum aggregations?

count - counts elasticsearch documents from the query to the index over time (every bucket_span)
sum - calculates the sum of a particular field within the documents over time (every bucket_span)

low_ and high_ variants allow you to control the reporting of anomalies only on the low-side or the high-side.