Andrew Ng cites cites the California ML paper you referred to:

* http://bit.ly/36r16vW

in his excellent 'The Batch':

* http://bit.ly/37AxBYF