how to fight your own biases when writing ML algorithms?

https://www.nature.com/articles/d41586-018-05469-3