Not where I live - unless you're studying for a Psych degree of course. Other BA students could take it as an elective. It was by far the most popular first year course in the Humanities department.
I don't know if it is a generalized feeling, but I sometimes tend to idolize my University years, where I was restlessly spending my time for pure pursuit of knowledge and truth (yeah, yeah :p). Really, those years were more about sweat and tears and curses about the bad teachers and the poorly worded manual and the absence of any documentation for the API provided to us (which was appropriate to face real life, I confess).
When I say I miss University, I miss the opportunity to choose classes at random and learn stuff I didn't know. I don't really miss the 8:30 classes and the long hours to produce lab assignments. I don't miss the effort, I miss the reward.
On a side note, maybe it is only their metrics about a class success and what "popular" is that are bad.
No reason not to just audit some classes at your local community college or check to see what classes are offered by your county.
The top-level offered might not be what you want, but they probably offer lots of diversity in choice. Plus it's stupid cheap.
I took a semester of pencil sketching from my county during a particularly stressful time, other than a pencil and some paper, all other materials were provided and it was one of the most relaxing things I ever did (plus I could feel the creative parts of my brain being exercised while I let the analytic parts rest for 3 hours a week). I think it ran under $200 and the class was only 12 people.
I was discussing this very issue with some friends of mine last night. The non-PhDs in the room could not believe what we were saying (core issues seems to be very limited opportunities in academia and research labs, a very strong job market for fresh graduates, and a general disdain for PhDs by startups).
Yes, I'm guilty of that too. The main reason I don't take more courses is the time it takes to take them. I've taken 3 courses (databases, algorithms I & II), and they were all great, but required a lot of work. So while I'm tempted to sign up for more, I need to have the time available to take them too. This is similar to buying books, but not having the time to read them (something I'm guilty of as well ;-))
I sign up for a lot of courses, but I drop maybe half of them a week in. Usually because it's more work than my casual interest in the subject warrants.
On the other hand, since these things started, I've finished 12 courses in a variety of fields.
I'm signed up for 5 classes starting in the next couple months, I'll probably finish maybe 3 of those. We'll see. :-)
Most of it in my case is learning for its own sake, but MOOCs are just about the best part of the internet for me. I took a Neural Networks class from Hinton and there's an upcoming class on Financial Markets from Robert Shiller. It's hard to beat that kind of access to teaching by absolute leaders in their fields.
I sign up for a lot of these, but am only interested in the material offered, not actually taking the course. Because of this, there are courses I never revisit, I just sign-up to save them for future reference.
my completion rate is somewhere around 30%, mainly because some courses are not what I expected to be, or I realize I'm not that interested in the subject.
The article is not saying why it is the most popular course. It only says that Ng is a personality and that ML is a hot topic these days. What is this different from other hot topics like trading, big data, or social networks analysis?
I just finished the "Thank You" video yesterday and, while I appreciate your humor (he does say both phrases a lot), I also want to point out what makes him so effective. It's a given that you have to be able to communicate your subject matter, but he's got an obvious enthusiasm for machine learning. And it's contagious - I'm a long way from Linear Algebra and Partial Derivatives (early '80s) but his easy-going style combined with how he presents the material makes it easy to follow along.
Now I just need a project where machine-learning would be an appropriate way of processing data!
I also finished the last of the videos over the weekend. I thought the course was excellent -- just the right amount of detail to get you started, and yes, his enthusiasm for the material is contagious.
Ng's machine learning is the first class on Coursera I actually saw through to completion, and I'm a little sad now that it's over. It reminded me of all of the best classes I took in college, how excited I was about the material and the inevitable letdown when the class ended.
Strange ... I felt the same feeling of sadness and it was only amplified by the last video where he spends half the time thanking us for taking the class.
How do I even express my gratitude?
This was my second Coursera class - I started with Martin Odersky's Introduction to Functional Programming in Scala (which I also highly recommend).
Apart from Prof. Ng's style and enthusiasm one of the things that makes specifically his ML class work is that he focuses a lot more on how to think about ML problems rather than the details of solving them.
I took the course without any CS background, but with some background in optimization techniques. And what i really took away was to model problems, how to optimize models, how to figure out if they are working at all, and most importantly the need to have different cross validation and test sets.
The last one is something that can be used for any statistical algorithm but often missed by new engineers.
Wasn't it one of the first course of this kind? (I'm talking about the first edition, since it was renewed recently) If so, that would explain a lot, as people were attracted to novelty.
I would assume it's because ML is just in the public's mind right now, particularly with the advancement in technology that is occurring (sensationalized by Kurzweil & other futurists of course)... Also, debate always seems to pop up regarding the Turing test and whether it is an accurate assessment of intelligence - with technology like voice recognition & siri (& siri-like) software approaching (broaching) Turing's declarations & theory of AI, I can understand a general curiosity from the masses from a philosophical POV. It does seem that the Turing test could soon be passed by a computer, but what that means (if AI is realized, or if Searle really defeated the logic of Turing...IMO, he did not) is something people will want to understand.... if they get bogged down by stats & CS, it seems apt; but the sign up rates & drop rates would merely indicate a demand for the philosophy regarding ML, AI, & current state of CS...if the courses offered are limited & people want to learn, they are going to get attention...
Statistician here. This Machine Learning is nothing but Statistics for me.
Also as a side note, if you are going to make something based on statistics you should consider to make it checked out by a statistician because it is such a big field that for example it takes 4 years to become a statistician.
Data science in 2013 has the feel of what software engineering should be, and possibly what it was before the MBAs got involved and attempted to commoditize development. It involves high autonomy and few non-technical people meddle in your work or hold strong opinions of how you "should" be doing the work. You generally get to pick your tools and set priorities.
The other major appeal of machine learning is that it touches all parts of computer science. You might have to go to a very low level (C, assembly, Cuda) for performance; but there's also tons of high-level work around expressing complexity elegantly-- hence the interest in using VHLLs like Python and Clojure for machine learning. Most professional software engineers are just munching tickets, but if you're in data science, you get to learn about databases, compilers, AI, information retrieval, and statistics at a more-than-superficial level.
When you have machine learning cred, you have a much better chance of being able to be an actual computer scientist instead of a cog in some dysfunctional CodeFactory.
1. Learn some ML. Either use Andrew Ng's course or Hastie's book or Bishop's. If you complete Hastie or Bishop you will be ahead (in theoretical knowledge) of 99% of professional data scientists.
2. Do some data science/machine learning work at your current job. It's not an either/or between SWE and DS. Software engineering is a huge part of real-world data science.
3A. Ask for a huge raise you won't get. When declined, say you'll take a regular cost-of-living raise if it comes with the title "machine learning engineer" (which, IMO, is more impressive than "data scientist").
-OR-
3B. Change jobs. After (1) and (2) you're more than qualified for a data scientist role.
It's best to do all of this when you're employed and have been at a job for at least 2 years.
Focus on quality rather than quantity. Be selective. You can send out hundreds of CVs in a night, but you have a finite amount of emotional energy.
Network, but the most useful thing you'll get out of connections is information, not good-ole-boy introductions. Go to as many Meetups related to your interests as you can. (Most cities have data science meetups, Scala and Clojure and Python meetups, et al).
Limit yourself to one coding test per week. They're not hard or time-consuming but they're emotionally draining.
Get a good night's sleep before the interview. If you're unemployed, resist temptations to drink or keep an unusual schedule. You need to be "on" at 9:00 am.
Keywords in job specs don't mean a whole lot. A great HR team doesn't mean a great company, and vice versa. People on HN say, "I wouldn't want to work for a company that wrote job specs like that". Well, in reality, there are a lot of good companies out there with crummy HR. So don't get too obsessed over keywords because most of what's in a job-specs ("looking for candidate with a track record") is non-information.
Thanks again! By the way, I wish I could find breakfast or lunch meetups. The last thing I want to do after work is head out to the city, deal with parking, etc.
Then most discover it's 90% history and statistics, with lots of debate about what's actually true, and don't take it any further.
Machine Learning is the CS equivalent - broad appeal, but to take it any further it's mostly maths and stats, which turns a lot of people off.