johnswentworth comments on Deep Learning Systems Are Not Less Interpretable Than Logic/​Probability/​Etc

[johnswentworth]

This is a really good summary, thankyou.