The Stanford AI class started today with the posting of a few introductory video instruction units. Most of these "Unit 1" vids were a camera on a writing pad making a few notes with a voice-over, and each concluded with a little "quiz" implemented as a javascript overlay on the video. On the plus side the videos are edited so there's not a lot of hemming and hawing (compared to the Khan Academy math lectures which are information packed but drag along as the presenter erases and re-writes his mistakes). On the minus side:
- The first set of quizzes were setup so as to lead into the next lesson and had nothing to do with what was covered in the actual video;
- The quiz answer system balked at about 2/3 to 3/4 of my responses and just refused to post them;
- The final set of videos and quizzes were concerned with an attempt to translate a Chinese Menu. If one already knows the ideograms one could tell them apart in the low rez video, but as an added insult the little quiz boxes obscured parts of the elements one was supposed to recognize and check off:
So I'm batting 62% on Chinese translation. Fortunately the inline quizzes don't count toward your grade. I just hope the real questions are not so well obscured.
Of a little more concern to me is:
- First -- The videos ended with these Introduction to AI bits that were mostly information free, even thought the schedule for the online class says day one also covers "Search" and has a homework assignment. In contrast the schedule for the real class has three days of lectures and a programming assignment(!?)
- Second -- The last Summary video listed the things that were covered. The last on the list was "Rationality" which was not mentioned in any of the lessons -- that I remember seeing. It is a key concept in their approach and is covered in some depth in Chapter 2 of the text book, where there are some probing exercise questions based on the definition.
No comments:
Post a Comment