> Scientists do not attempt to decipher the “black box,” or the opaque processes that the system undertakes, as long as they receive the outputs they are looki

> Scientists do not attempt to decipher the “black box,” or the opaque processes that the system undertakes, as long as they receive the outputs they are looking for. This seems a bit disingenuous as there's a whole research field (explainability) that does exactly that. This is not the widest of the whole machine learning domain, but it's there and there's significant work around it...