4 ms·
I'm not seeing an actual statistics course in there. Also GOFAI techniques seem to be only briefly mentioned in an "Introduction to AI" course, which just doesn
by zozbot234 6y ago
I'm not seeing an actual statistics course in there. Also GOFAI techniques seem to be only briefly mentioned in an "Introduction to AI" course, which just doesn't cut it in my view - you'd need something like an Operations Research class to really explore that stuff in more detail. Some of the more focused CS content could be cut to make room for this; OS's and DB's has been mentioned already, but maybe one could also remove the Compilers course.
- disgruntledphd2 6y agoYeah, I'd probably remove compilers, and add classes on statistics, experimentation and actual data analysis. I kinda find it hilarious that there is no mention of analysing data in this course, given that that is a core skill that all data professionals actually need. I mean, where do they think the models come from? There's also little to nothing about ETL, which is pretty much all you do in most data sciencey jobs. I actually think that this course is pretty revealing of the worldview of a lot of AI researchers, and may be why predictions of AI's imminent dominance across multiple fields appear to be not particularly accurate.
- zozbot234 6y ago> "There's also little to nothing about ETL, which is pretty much all you do in most data sciencey jobs." Maybe one could design a custom "Databases and ETL" class for this, to make it useful for more than just data-sciencey jobs. ETL is usually covered to some extent in stats classes though, AIUI.