Slashdot Mirror

What is he actually talking about? Is there actually programs here in the US that offer an 18 month course that promise to make someone a "machine learning scientist", whatever that is?

There's a 12-month boot camp/nanodegree program for machine learning. That plus six months of on the job training should do the trick.

https://www.udacity.com/course/machine-learning-engineer-nanodegree--nd009

Uh, it's available for free to Americans too, at least to the ones who can use a search engine: https://www.udacity.com/course...

Yes it's available in the US, and it's in English btw so no excuses. I am amazed you cant google for 5 seconds to find it, and then you complain that a dude who has never seen a computer in his life can compete with you? https://www.udacity.com/course...

And yes it's a FREE course.

If you want to start playing with some deep learning models, I would highly recommend this page. It provides some basic examples that run right in your browser. Also, this page provides a great guide to working with neural networks without getting bogged down in a bunch of mathematical equations.

Another great resource is Caffe. Caffe is a deep learning framework that will let you define a wide variety of neural networks by just writing a text file. You can run Caffe applications in CPU or GPU mode (a lot of open source deep learning code will only work with GPUs, so being able to run things either way is a nice feature).

If you want to do computer vision, make sure that you read up on fully convolutional neural networks, because they are the big thing right now.

Remember that story about a program that was able to learn how to play just about any Atari game? That is called reinforcement learning, and that's a big thing right now too. Udacity has a great course on reinforcement learning.

If you really want to be employable then I recommend getting a nanodegree in Machine Learning through Udacity. https://www.udacity.com/course... One of the things that makes their program stand out is that they use project based learning. Upon completion you have projects that you can present to perspective employers.

Give it some time.

As any AI researcher will tell you, we know how the brain works and Geoffrey Hinton's recent paper is nothing short of a breakthrough, and will lead to us having strong AI programs real soon.

We have IBM's Watson, a program that actually understands the information it's processing and will be used to augment medical diagnosis, SIRI, a personal assistant application that actually learns, and MAKO, a program who can do anything on a PC!

IBM is already making neural network chips that implement the way the brain really works, a program the learns the same way that a child learns, and many, many more!

We have courses that teach you AI, and ... it's easy!

Give it some time! We need to let the AI mature like a fine wine, and filter down into consumer devices.

It's coming soon - it really is!

Udacity has a free class on Git and GitHub. I recommend it. They spend a little too much time on writing a chart that diagrams the different parts of Git, but the class is well-structured and clear.

People sign up and never finish because the courses are downright awful. And there's no mind nor incentive for them to get better. Instructors think that just recording a lecture and putting it online is good education, but it isn't.

Watch Daphne Koller droning on about graphical models as the video shows her standing at a lectern talking, or showing a powerpoint-style frame while she reads the text on the frame to us.

Watch Anant Agarwal go through a *hugely* dense and boring derivation, only to stop before the end and say "but this derivation is too hard, there's an easier way". Twice. For the same result.

Try to figure out how many degrees of freedom a soccer ball has, then argue with Sebastian Thrun because the answer he thought you should have entered is not the mathematically correct one. (Also, see if you can figure out what this has to do with AI.)

For a breath of fresh air, watch Donald Sadoway take you through a delightful and satisfying explanation of chemistry. (Ignore the 1st lecture which is about class scheduling.) It's wonderful.

I could cite two dozen *major* problems with selected online courses - things that go counter to the fundamental goal of learning that would be obvious to someone familiar with human learning mechanisms or a testing group or even a member of Toastmasters. When I point these out to the chief scientist at edX, he responds with "we can't change the way we do things because of X".

Let me repeat that: the *chief scientist* at edX has no control over teaching techniques or video methods or course quality.

Some people (ie - Dr. Sadoway in the link above) have figured out how to do it right, but the vast majority aren't interested in quality. It's unfortunate that edX got all those millions in seed money, because we'll have to wait until they burn through it before they get hungry enough to worry about quality and effectiveness.

It's insane.

You can find a lot of open CS courses from prominent universities offered online with lecture videos, assignments, projects, the works:

edX
udacity
coursera

Some offer certificates, but most universities won't accept these. You can try to get the silly credits like English requirement done at a community college which will offer night classes. If you can't give up your 9 to 5 then you can attend a state school or community college part time. Some employers partner with state/community colleges for internships and jobs such as Lone Star College and HP (which actually share a campus in northwest Houston!).

Udacity announced in November for-pay MOOC classes: http://blog.udacity.com/2013/11/udacity-innovation-is-in-our-dna.html /K

I asked one of the edX higher-ups (not Dr. Agarwal) about this.

Apparently, there's a lot of interest from companies looking for good talent. For example, people who score in the upper 10% of a high-tech course would be of interest to many companies. Especially in today's market, where putting out a job listing will get thousands of inappropriate resumes.

IIRC, the top 2% of the original AI course (Udacity, not edX) students could optionally have their resume sent to Google for consideration.

The edX higher-up was of the opinion that the future of high-tech hiring would be in the form of online course grades - you would list the courses you took and the grades you got as part of your resume. Companies would advertize for people who had taken specific courses and received certain grades.

I took that course: https://www.coursera.org/course/hetero

I also took a course from Udacity: https://www.udacity.com/course/cs344 but this one I didn't finish, I've done perhaps 30% of it (I already had finished Coursera's). One of these days I'll go there to close matters :-)

The courses in Udacity are "always online", so anyone can register anytime and finish the course with his/hers own pace. Quizzes, exams and grading with certificate included have no fixed limits. On the other hand, the courses from Coursera have deadlines and run more or less in parallel with "snail" university schedules, with start and stop dates, with time limits in quizzes and exams, etc. (You can usually see videos, and do quizzes anytime after they end, but no certificates and grading AFAIK).

Both courses were good -- I recommend both, -- we did homeworks in Amazon's cloud transparently, and certainly both were "sponsored" by Nvidia, coz we learned only CUDA. (Perhaps there was a brief blah blah about competing alternatives.)

But from what I've seen, if someone is afraid from CUDA, then its better to run away very fast from alternatives (OpenCL) :-)

So there's nothing really easy about GPU programming. You can look at C++ AMP from Microsoft, OpenMP or one of the other abstractions but you really need to understand how these massively parallel machines work. It's possible to write some perfectly valid code in any of these environments which will run SLOWER than on the CPU because you didn't understand fundamentally how GPUs excel at processing.

Udacity currently has a fairly decent intro course on GPU programming at: https://www.udacity.com/course/cs344

It's based around NVIDIA and CUDA but most of the concepts in the course can be applied to OpenCL or another GPU programming API with a little syntax translation. Also you can do everything for the course in your web-browser and you don't need an NVIDIA GPU to finish the course exercises.

I'd suggest running through that and then deciding on what API you want to end up using.

Check out the Udacity class on parallel programming. It's mostly CUDA (I believe it's taught by NVIDIA engineers): https://www.udacity.com/course/cs344

CUDA is generally easier to program than OpenCL. Of course, CUDA only runs on NVIDIA GPUs though.

We're really living in amazing times.

Most online courses to date have been lacking in one aspect or another, most notably student interest - drop rates of over 95% are common. Teething pains probably, as teachers begin to recognize that a) courses online must be presented in a different way, and b) teaching techniques must be effective (in terms of keeping student interest) when the audience is not captive.

Recently I saw this gem, which is extremely good. Good presentation, good technical quality (web form scoring &c), good content, and some experimental techniques in keeping student interest.

While I don't like the techniques used for keeping student interest in this course, they are at least experimenting with new techniques and learning from past mistakes. The quality keeps getting better.

Their business model varies, but one site hopes to provide an MBA ensemble for $50 (Udacity) and another gets finders fees from companies that hire the top scorers (edX). And of course there's Kahn academy, which is turning high-school education upside down.

In a couple of years, you will probably be able to get a complete high-quality education by self-study over the internet for thin money. You'll be able to study as much as you want for whatever topic you want and for as long as you want.

No more massive student loans just to get a decent education.

Another example of a moribund business model being overtaken by new technology.

Amazing times indeed.

Udacity and Coursera both offer free online college-level courses in programming. Udacity's focus is primarily on Python (at least in the courses I've taken), but it looks like Coursera's offerings include C++. Any of these should give him a good start in object-oriented programming.

I take the last part back. They apparently have started offering optional proctored exams for some of their online classes.

I would recommend Udacity's CS 101 course along with LPTHW .

The thing I liked about that Udacity course is it is study at your own pace. You can do as little or as much as you want. On top of that you are building a web crawler throughout all of the Units.

I'd also recommend code academy since they added a lot more python modules. It would be a good way to reinforce ideas he is learning whatever course he is going through.

The good thing about LPTHW over the others is you are also setting up a programming environment on your own computer rather than using someone elses interpreter.

"The current learning language for Computer Science is Java."

You may be a little older and behind the current trends. All quality schools I'm aware of teaching CS classes are using Python for several years. Additionally, I think Python surpassed Java for introductory course to 6th graders a couple of years ago.

http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-00sc-introduction-to-computer-science-and-programming-spring-2011/index.htm

http://www.udacity.com/

"The language of choice for Linux/Unix system administration is Perl."
I suspect Python has far surpassed Perl use in general management... Maybe looking at Red Hat or Ubuntu might help with some info.

How about the numerous project in Python compared to Perl?

If you choose to DIY, it may interest you to take a look at Steve Huffman's (Reddit) at Udacity : CS253 How to Build a Blog. Sure, a blog is not a forum but they share some key elements. Steve uses Google App Engine as the server and codes in Python and Jinja2.

I think what we'll see is that Google cars will be uploading both human and computer-driven data to a service. This service will continually fine-tune the maps - so it can identify that there has been a stationary object in the same place the last 6 times (it's probably a mailbox, street sign, or light pole). So streets that weren't mapped before, or maybe didn't have enough data to be considered computer-safe, could become so rather quickly.

If you're into statistics at least a little bit and have an interest, I strongly suggest Udacity's course on self-driving cars. It gives a lot of insight into how the car works.

Steve Huffman, one of the creators of Reddit, talks about this exact solution during his Udacity class, Web Application Engineering. http://www.udacity.com/overview/Course/cs253/CourseRev/apr2012 I think it was during week 4 "Whom to Trust," but I don't have links to the exact video. So in short, yes, it has been done effectively in the past, though I believe they wrote their own code to do it.

I don't know if anyone has been keeping track, but there's this thing called the internet where you can get a really good education for free. Tablets will give children access to this internet.

We currently have four major players in this arena:

• Khan academy, for high-school up through 1st year college
• Coursera, college level
• MITx/edx, college level
• Udacity, college level

This is in addition to all the universities which are putting lecture videos online, along with course materials and (in a few cases) the textbook content. Oh, and youtube videos of lectures, and the zillion-and-one websites explaining whichever subject you're interested in. Google "relativity" or "tensors" sometime - see if you can find an explanation that works for you.

An experiment in India has shown that when you give uneducated, poor children access to an internet-connected computer they figure things out on their own. Complex, interesting, and difficult things that you might not expect an ignorant user to manage. (Such as typing a thank-you note without access to a keyboard.)

This is all you need, kids will figure things out for themselves. Having a teacher to nudge them in the right direction, or help them over a difficult part is just gravy.

Kids are voracious learners, and have always been. Abe Lincoln used to sit at home practicing his "ciphering" (arithmetic) by drawing numbers on a shovel with charcoal. Over and over, until he got comfortable with the math. All kids do this - it's in the nature of growing up.

Just giving kids access to material will be a huge leap over the current situation. Schools and teachers are extra.

Intro to Statistics: Making Decisions Based on Data

We're moving from a culture that encourages individual learning/mastery/understanding of the things used in life, to one of apathetic dependence on convenient 'service'. This is intellectually stunting, which causes all kinds of other problems.

And yet, there's this.

Get them to go to http://www.udacity.com/ and take the CS101 course. It'll teach them python and the general concepts of programming without being too difficult and it also takes the pressure out of you (or someone else) teaching them ... and at the end they can get a signed certificate to say they did the course (it holds no weight for getting a job, but at least it's a nice piece of paper). :-)

Have a take an introductory course. http://www.udacity.com/overview/Course/cs101/CourseRev/apr2012

What Udacity as well - they are quickly ramping up their courses. New one for physics starting June 25.

Udacity has a physics course it's rolling out at the end of the month. Looks fairly basic, but you'll have to decide for yourself if the level is appropriate for you.

And there seem to be some introductory courses from both Stanford's Coursera and Udacity (Sebatian Thrun's new start up).

They're programming oriented, if I'm not mistaken, though.

Have him learn python. On any OS.

If you are going to teach him Python, have him take CS101 at Udacity. It is more fun than reading a book.

Better yet. Take his free online class. http://www.udacity.com/

In case anyone is interested, Udacity will be offering CS 373: PROGRAMMING A ROBOTIC CAR again starting in April. The first class is finishing up and Sebastian Thrun does an excellent job demonstrating the concepts required to make the self-driving car happen.

As a geek and a person currently taking Sebastian Thrun's Building a Robotic Car class, I think the self-driving car is the most interesting project since the invention of the hyperlink.

For Google to fund efforts like this goes a long way with me in the PR department.