I've been spending the past hour on "Leaning how to learn" because of the recommendations here. I was... well, the course was definitely not quite what I expected.
What I'm seeing here is a bunch of "tricks", and a lot of "brain facts" which I would usually dismiss as pseudoscience. The course almost feels like a scam. What gives?
All of the course content is based on the hard science, but is simplified so that every one can understand. You can look at the background of one of the course instructor:
https://www.coursera.org/instructor/terry
I didn't think the R course was that great. I only did the first course in the speciality but I thought the assignments didn't match the lectures very well.
I don't really know R but I still got through OK (100%) but it didn't compare well with the EdX AMPLab Spark course I did around the same time.
+1 for Dan's crypto1. I don't think crypto2 has been taught yet via coursera, as I've been waiting to take it and have seen it pushed back several times, and I've seen others say they'd been watching it get delayed for years. It does seem that Dan has been recording videos for part 2 in 2015 though (according to one of his students), so there's reason for hope that it might happen in 2016.
I think we'll know in a couple of weeks if its going to start on Jan 11th or not. If its not ready yet, it'll be pushed back by a month or two by mid December. That's how the previous postponements were done.
I finished Cryptography I in March 2012 and wanted to take Cryptography II ever since. Every time the announced time is close it gets pushed back by 4 months.
Programming Languages is one of those courses that just keeps on giving.
A great basis for functional and Lisp fundamentals. I'm just starting a journey into Erlang and that course has meant that the switch isn't as difficult as it could have been.
The course format was interesting. I'm not 100% on board with doing peer assessment, but I did like being able to see how other people handled the assignments.
Convex Optimization by Stephen Boyd (Stanford EE364A) available on itunesU. There's also a CVX101 Mooc[1], but I don't how it's different from the original material. IMO it's not the topic itself, but the invaluable material for machine learning, statistics and applied mathematics. And Boyd has such a huge insight on the topic it's always a pleasure to watch his lectures.
Thanks for sharing! On iTunes U are there assignments or other course materials, or just the video lectures? I only see the videos there and I'd like to get my hands dirty practicing assignments and not just watch the lectures.
Then I suggest the website of the course where you can find homework and lots of material. You can follow the course as if you were enrolled. Also the free pdf copy of the book will certainly be useful.
CS188.1x Artificial Intelligence by BerkeleyX at edx.org. [1] I took this course back in Spring 2013 and I really enjoy the course project of making an intelligent Pac-Man. :) Through this course, besides learning AI, I also learned Python (before this, I didn't know how to code in Python at all). And with the knowledge from this course, I made a simple connect four game with AI implementation as the player's opponent. [2]
Taking this course right now. Just about to start Homework 2. Awesome stuff. I've laid out a curriculum for myself starting from this course and ending in ML expertise. :D
Discrete Optimization https://www.coursera.org/course/optimization -- I really enjoyed this challenging class with a very dynamic teacher, and organized around a set of tough problems that you can tackle using a choice of optimization paradigms (e.g. you can decide to "specialize" in "local search" if you want, and try to solve eveything with it).
I haven't taken many online courses, but here are my favorites (notice a trend):
Andrew Ng's ML Class - This makes the list because it is incredibly useful. I didn't have much background in the field and this class is a practical survey of ideas. Not a ton of depth, but exposes you to a lot of information gently.
Daphne Koller's PGM Class - This was the most rewarding. I banged my head on a lot of this material, but it was an incredible feeling when things started to click. That I was able to complete this class is a testament to Dr. Koller's excellence as an educator.
Dan Jurafsky's and Christopher Manning's NLP Class - This class was the most fun. I thought the exercises were incredibly well designed. Unlike the first two courses, the exercises were a lot more interesting. For ML and PGM, you mostly know when you have the answer and you are rewarded with 100%. NLP assignments are based on how well your system generalizes, which made me try harder to improve my systems, and helped me enjoy the course.
NLP has been on my watchlist for a semester now. No word on when future sessions will happen. I'm about to give up on that and just watch the old videos. Too bad I won't get the assignments :/
Koller's PGM was fun but very hard to generalize outside of the problems presented in the class. Opportunities to implement loopy belief propagation just don't seem to come up as much ad I'd like.
That's because most were toy examples dealing with discrete distributions in a few dimensions. Granted, these are mathematically easier to deal with, but not representative of real-world scenarios.
Wow, that PGM course looks exactly like what I need to get my head around GMs! I'm using Andrew Ng's class to support my learning for my MSc in Data Science, and it's been an invaluable resource. There is nothing like hearing the same material explained by different people to really hammer the topics home, I'm looking forward to getting stuck into that PGM course.
I've only just started it, but this looks great in terms of the content they cover, and they also provide quite a few programming tasks.
https://class.coursera.org/hiddenmessages-006
It's the only MOOC I've taken that was anywhere close to the kind of experience I had as an actual undergraduate at MIT. Outstanding lectures with accompanying lecture notes, challenging but rewarding problem sets, lots of interaction by the professor and other staff in the forums.
It's not a course in the sense of having problem sets and grades, but V. Balakrishnan's lecture series on classical physics (https://www.youtube.com/playlist?list=PL5E4E56893588CBA8) is amazing, just incredibly dense with insight.
I haven't taken many online courses but my wife & I just took the course "Learning How to Learn" (https://www.coursera.org/learn/learning-how-to-learn) together and I wish this was available before I went to university. They do a great job of presenting the content and provide a lot of references for additional reading for those that have a deeper interest. It should probably be considered the pre-requisite to all other online courses!
Introduction to Cryptography, by Christof Paar. His book '
Understanding Cryptography: A Textbook for Students and Practitioners' also provides great insight to the subject. https://www.youtube.com/watch?v=2aHkqB2-46k
I took these to prepare for first-job interviews coming out of grad school. Got an offer from a company frequently mentioned on this site, so I guess they helped.
The Harvard computer sci course really helped me out. I didn't take the class, but put all classes on my iPod. I would listen to lectures while exercising at night.
This course really helped me understand the ever changing computer lingo. I probally should have done the lessons.
Once you get used to the vocabulary, and all the acronyms--it's all starts to fall into place.
166 comments
[ 2.5 ms ] story [ 219 ms ] threadLearning how to learn is the best course online that any one can take.
and with the class here -
http://datagrad.blogspot.com/2013/01/my-most-recent-mooc-exp...
Even with some R experience I didn't feel like I got enough information from the R Prog lectures to complete the assignments.
What I'm seeing here is a bunch of "tricks", and a lot of "brain facts" which I would usually dismiss as pseudoscience. The course almost feels like a scam. What gives?
I don't really know R but I still got through OK (100%) but it didn't compare well with the EdX AMPLab Spark course I did around the same time.
(Part 1) https://www.coursera.org/course/crypto (Part 2) https://www.coursera.org/course/crypto2
I also liked Udacity crypto course, less formal but with great "hands on" exercises:
https://www.udacity.com/course/applied-cryptography--cs387
"Compilers" - https://www.coursera.org/course/compilers
A great basis for functional and Lisp fundamentals. I'm just starting a journey into Erlang and that course has meant that the switch isn't as difficult as it could have been.
The course format was interesting. I'm not 100% on board with doing peer assessment, but I did like being able to see how other people handled the assignments.
This course is amazing, especially for the assignments.
[1]https://lagunita.stanford.edu/courses/Engineering/CVX101/Win...
http://stanford.edu/class/ee364a/index.html http://stanford.edu/~boyd/cvxbook/
Also, intro to comp sci by Harvard's open courseware. Without these, I might've dropped out of comp sci in my second year [2]
[1] https://www.youtube.com/watch?v=kBdfcR-8hEY [2] https://www.youtube.com/watch?v=z-OxzIC6pic&list=PLvJoKWRPIu...
[1] https://www.edx.org/course/artificial-intelligence-uc-berkel...
[2] http://kenrick95.github.io/c4/demo/
Made by the guys from The Blue Bottle, splendid tutorial!
Andrew Ng's ML Class - This makes the list because it is incredibly useful. I didn't have much background in the field and this class is a practical survey of ideas. Not a ton of depth, but exposes you to a lot of information gently.
Daphne Koller's PGM Class - This was the most rewarding. I banged my head on a lot of this material, but it was an incredible feeling when things started to click. That I was able to complete this class is a testament to Dr. Koller's excellence as an educator.
Dan Jurafsky's and Christopher Manning's NLP Class - This class was the most fun. I thought the exercises were incredibly well designed. Unlike the first two courses, the exercises were a lot more interesting. For ML and PGM, you mostly know when you have the answer and you are rewarded with 100%. NLP assignments are based on how well your system generalizes, which made me try harder to improve my systems, and helped me enjoy the course.
Andrew Ng's ML Class - https://www.coursera.org/learn/machine-learning
Daphne Koller's PGM Class - https://www.coursera.org/course/pgm
Dan Jurafsky's and Christopher Manning's NLP Class - https://www.coursera.org/course/nlp
Thanks for these!
[+] http://wayback.archive-it.org/3671/20150529001651/https://ww...
It's the only MOOC I've taken that was anywhere close to the kind of experience I had as an actual undergraduate at MIT. Outstanding lectures with accompanying lecture notes, challenging but rewarding problem sets, lots of interaction by the professor and other staff in the forums.
It's not a course in the sense of having problem sets and grades, but V. Balakrishnan's lecture series on classical physics (https://www.youtube.com/playlist?list=PL5E4E56893588CBA8) is amazing, just incredibly dense with insight.
I took these to prepare for first-job interviews coming out of grad school. Got an offer from a company frequently mentioned on this site, so I guess they helped.
https://www.edx.org/course/introduction-computer-science-har...
This course really helped me understand the ever changing computer lingo. I probally should have done the lessons.
Once you get used to the vocabulary, and all the acronyms--it's all starts to fall into place.