Ask HN: Recommend me a course on Coursera • Hacker News Board

Re: Ask HN: Recommend me a course on Coursera

#91

Post by tasubotadas » Fri, Apr 10, 2020, 9:20 AM UTC

bcrosby95 wrote: post #76

I took it a while ago, and it was a ton of work, but I really liked https://www.coursera.org/specializations/probabilistic-graph...

I took it relatively recently but I have mixed feelings about it. While the topic itself is very interesting and they have some interesting exercises, the explanations in the videos are seriously lacking.

There are big jumps in reasoning and explanations so you'll need to rely on external material quite a bit (I've used Bayesian Reasoning by Barber) Also, exercises are presented in Matlab/Octave which makes them a pain to work with.

After completing the specialization, I've almost felt that you could just do a course that explains this course.

Re: Ask HN: Recommend me a course on Coursera

#92

Post by leandron » Fri, Apr 10, 2020, 9:22 AM UTC

sonabinu wrote: post #64

I would start with 'Learning how to Learn' - https://www.coursera.org/learn/learning-how-to-learn

I took it a few years back. Highly recommended.

Re: Ask HN: Recommend me a course on Coursera

#93

Post by tasubotadas » Fri, Apr 10, 2020, 9:24 AM UTC

silverdrake11 wrote: post #45

airstrike wrote: post #7

https://www.coursera.org/learn/machine-learning with Andrew Ng

I would recommend Andrew Ng's updated course on Deep Learning with python instead. https://www.coursera.org/specializations/deep-learning

Yeah - the updated version is much better (I've completed both of them) just because you don't need to struggle with Matlab.

Overall, this course is extremely good mostly because Ng covers the essential theoretical topics and gives some practical advice. Also, the topics are explained really well and you do not need to look up additional material. Also, I really appreciate that he took the time to derive those equations while others just drop the results.

Re: Ask HN: Recommend me a course on Coursera

#94

Post by tasubotadas » Fri, Apr 10, 2020, 9:27 AM UTC

smabie wrote: post #23

Let me recommend you an udacity course instead. This is hands down the best course I've ever taken in my life: AI for Trading https://www.udacity.com/course/ai-for-trading--nd880 Includes an introduction to finance/markets, and goes into strategies, multi-factor models, and deep learning. Great projects too!

I've taken it but I wouldn't recommend it - it was pretty shallow. That guy used to have a more complete version of the course on Coursera. I believe that it became a specialization.

Re: Ask HN: Recommend me a course on Coursera

#95

Post by tasubotadas » Fri, Apr 10, 2020, 9:37 AM UTC

master_yoda_1 wrote: post #12

most of the courses on coursera are for uninitiated and have very shallow content (there are some rare exception). So if you like you can search for a good book or some video lecture from good university. like this is for statistical learning " https://online.stanford.edu/courses/sohs-ystatslearning-stat...

That's true. I think most of the recent courses from John Hopkins university illustrate this the best. They even have a separate course for Linear Regression!

However, if you spend enough time sifting through the junk, there are some decent material.

Re: Ask HN: Recommend me a course on Coursera

#96

Post by DrNuke » Fri, Apr 10, 2020, 10:00 AM UTC

Ok but do you need a course meant as a vocational, subject-specific training or something more liberal that helps you think better for the long run? I would go for the latter.

Re: Ask HN: Recommend me a course on Coursera

#97

Post by hawkjo » Fri, Apr 10, 2020, 10:14 AM UTC

betamaxthetape wrote: post #89

Build a Modern Computer from First Principles: From Nand to Tetris (Project-Centered Course) https://www.coursera.org/learn/build-a-computer and then part 2: https://www.coursera.org/learn/nand2tetris2 [Both courses are free] These are fantastic courses, by far the best MOOCs I have ever taken. I went into them knowing nothing about computer architecture, and by the end of the first course I was able to design a full…

I have to also mention nandgame.com here. Similar idea, but all presented as an interactive, browser-based game. Start with nand gates and build up to a functioning computer. You never get to tetris, but it's great and super easy to start playing/learning. Delightful.

Re: Ask HN: Recommend me a course on Coursera

#98

Post by exdsq » Fri, Apr 10, 2020, 10:33 AM UTC

I’ve started working through the Verification series by EIT which have been challenging and interesting. If you’re interested in formal verification, check them out.

Re: Ask HN: Recommend me a course on Coursera

#99

Post by kkarakk » Fri, Apr 10, 2020, 10:44 AM UTC

Dowwie wrote: post #88

Coursera is great but you're going to miss a lot of opportunities by limiting yourself to just that platform. For instance, it hasn't been widely advertised, but you can essentially take Steven Pinker's 2020 course at Harvard on Rationale: https://stevenpinker.com/classes/rationality-gened-1066 There's upwards of 20 hours of video on this course alone. You don't get that kind of depth from most Coursera MOOCs. Furthe…

Seeing as how the course is hosted on a proprietary video hosting service called Panopto that is a SaaS platform i think the chance of this lecture surviving for even 5 years is vanishingly small

Atleast coursera courses are easy to rip for later use...

Re: Ask HN: Recommend me a course on Coursera

#100

Post by Eugeleo » Fri, Apr 10, 2020, 11:18 AM UTC

Dowwie wrote: post #88

Coursera is great but you're going to miss a lot of opportunities by limiting yourself to just that platform. For instance, it hasn't been widely advertised, but you can essentially take Steven Pinker's 2020 course at Harvard on Rationale: https://stevenpinker.com/classes/rationality-gened-1066 There's upwards of 20 hours of video on this course alone. You don't get that kind of depth from most Coursera MOOCs. Furthe…

Looks great! Is this the rationality often mentioned (and propagated) on LessWrong, Slate Star Codex and similar?