Many monitoring tools like AWS Cloudwatch/ DataDog/ New Relic and others provide some sort of anomaly detection option? For example - Cloudwatch introduced this(https://aws.amazon.com/blogs/aws/new-amazon-cloudwatch-anomaly-detection/) in October last year. DataDog has outlier detection (https://www.datadoghq.com/blog/outlier-detection-algorithms-at-datadog/) algos which one can use.

Do you folks use these features? Why or why not?

If yes, what is the key improvement - does it improve the quality of alerts or reduces false positives?