3 ms·
Corporate Learning at Scale: Lessons from a Large Online Course at Google [pdf]
- super_sloth 12y agoThat sounded like a fun course. Has anyone taken that and say Andrew Ng's or Yaser's courses? How did they compare?
- wodenokoto 12y agoI was expecting them to go a bit more into details about how many actually completed the course, if ways of taking the course had effect on completion and even to look at code check-in behavior after the course. Instead the only results where survey and they didn't even mention how many completed it. The following paragraph sums up ALL results in the paper: > Forty-six percent of post-class survey respondents are “plan ning to use machine learning as a result of this class,” and six percent report that they are already “using machine learning as a result of this class.” Of final survey respondents,62% report having machine learning conversations with others.
- modzu 12y agolol for real.. title is a bit misleading one thing i did find interesting is their intent/ability to evaluate the effectiveness of the course by the subsequent use of the tools, rather than traditional assessments. i know ive aced exams only to forget everything the next day, and conversely there are courses i nearly failed or didnt finish that nevertheless had a profound impact on my learning. i dont think this is something that is really measured or tracked in traditional education
- ErikBjare 12y agoThis course haven't had the chance to leak out into the open web has it? Someone from Google who has access please do, it's in the best interest of everyone. To be honest I'm kind of surprised they didn't release it already. As someone who has completed Andrew Ng's course it would be interesting to compare. > The lecture was followed by one or more case studies, whereexperts explained how the techniques taught in lecture hadbeen applied to solve important problems at Google. That would certainly be interesting to watch.
- eginhard 12y agoConsidering that it makes use of internal machine learning tools and probably refers to a lot of other internal technologies it wouldn't be of much value to anyone outside Google I guess.