The irony is that only a tiny fraction of the code in ML/AI systems is devoted to learning or prediction. Rather, the majority of complexity manifests in data preparation, feature engineering, and operationalizing distributed infrastructure reqd to perform these tasks at scale

The irony is that only a tiny fraction of the code in ML/AI systems is devoted to learning or prediction. Rather, the majority of complexity manifests in data preparation, feature engineering, and operationalizing distributed infrastructure reqd to perform these tasks at scale

— Nikhil Kulkarni (@kulkarninikhil) November 29, 2017

November 29, 2017 at 09:17AM
via IFTTT