LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   General comments on the course (http://book.caltech.edu/bookforum/forumdisplay.php?f=140)
-   -   Prerequisites for the course? (http://book.caltech.edu/bookforum/showthread.php?t=4492)

Tellur 08-14-2014 04:25 AM

Prerequisites for the course?
 
Hi,
I'm pleased to see this course offered on edX again. But unfortunately I'm very much in doubt as to whether my current background knowledge will suffice to benefit from this course. I already wanted to take Andre NG's course on coursera in June but immediately backed down when I saw that previous knowledge of hypothesis testing and the least squared method was required. Though I have to admit that I didn't even try to fill in these maybe rather small knowledge gaps due to my limited time then.

My current background knowledge is:
- a reasonable good command of high school level calculus (just refreshing it)
- familiarity with basic arithmetic matrix operations (not more than high school level)
- very basic Python programming experience (currently taking the introduction course on udacity)
- only VERY basic familiarity with concepts of probability and statistics

The least mentioned is definitely my weakest point. I'm only superficially familiar with the very basic concepts like permutation/combination/mathematical expectation/mean/median/mode/standard deviation/Bayes' Theorem/standard deviation and variance/gaussian distribution and I would have to refresh even these basic concepts.
I have no experience with linear regression or the method of least squares. Though I do understand both concepts (without knowing the exact formulae) I've never really learned or used them.
I have no experience at all of hypothesis testing.

Given this "modest" background knowledge, will a few weeks (let's say 10-15 hours/week) of preparation to fill the gaps suffice to be really adequately prepared for this course?

In case yes, my most important question is: on which subjects should i concentrate in my preparation?

(In case no, I would nontheless try to use the archived version of the course at a later point after thouroughly filling in the gaps. So in any case I would be interested in the question on which subjects I should concentrate to prepare for this course.)

yaser 08-14-2014 09:45 AM

Re: Prerequisites for the course?
 
I think it is quite doable given your background. There are resources on the web, such as Khan academy videos (I know of them by reputation, but not in detail) and even wikipedia, that can help with further preparation.

I would suggest, between now and the start of the course on edX, that you look at Lectures 1-3 of the course on youtube or iTunes. You can then look up background material at a relaxed pace to make you understand these lectures well. They involve probability, matrices and calculus. You can also try to program some of the equations presented in these lectures and perhaps plot some results to warm up your programming skills.This way, you will acquire targeted background, and also get a head start on the course so that you are not pressured by basic prerequisites in the heat of the battle.

Good luck. Please let us know if a particular resource was useful for this background so others in a similar situation can benefit from it.

Tellur 08-16-2014 03:13 AM

Re: Prerequisites for the course?
 
Thank you very much indeed for your answer. I'll follow your advice and look right into the first lectures.
I'm already using (amongst others) Khan Academy to refresh up my calculus background. One of Khan Academy's distinctive feature is the short length of the videos. The subjects are divided into very small videos of mostly between 5-15 minutes. So it's easy to get oneself to learn even if one is not in the mood or doesn't have the time to learn an entire hour and longer.

Especially for Calculus and Linear Algebra I also highly recommend Patrick JMT. Though these Videos can unfortunatley only be watched online and not be downloaded.

What I'm still looking for is an entry level probability and statistics course that emphasizes strongly on those subjects which are particularly important for machine learning and data mining. So far I got the impressions that these particular elements of probability and statistics (as far as I can identify them) are quite scattered between different courses or series of consecutive courses which are bases on each other, so that it'd be very time consuming to use them for preparation.
Should I find such a course I promise to post it here.

shameer 09-18-2015 05:08 AM

Re: Prerequisites for the course?
 
MIT had an excellent introductory course on probability which was offered on edx. Lectures might still be available in the archived version

JennPendic 07-11-2017 11:36 PM

Re: Prerequisites for the course?
 
Quote:

Originally Posted by Tellur (Post 11713)
Thank you very much indeed for your answer. I'll follow your advice and look right into the first lectures.
I'm already using (amongst others) Khan Academy to refresh up my calculus background. One of Khan Academy's distinctive feature is the short length of the videos. The subjects are divided into very small videos of mostly between 5-15 minutes. So it's easy to get oneself to learn even if one is not in the mood or doesn't have the time to learn an entire hour and longer.

Especially for Calculus and Linear Algebra I also highly recommend Patrick JMT. Though these Videos can unfortunatley only be watched online and not be downloaded.

What I'm still looking for is an entry level probability and statistics course that emphasizes strongly on those subjects which are particularly important for machine learning and data mining. So far I got the impressions that these particular elements of probability and statistics (as far as I can identify them) are quite scattered between different courses or series of consecutive courses which are bases on each other, so that it'd be very time consuming to use them for preparation.
Should I find such a course I promise to post it here.

hi, I am in similar shoes - just wondering if there are new recommendations since. I think PatrickJMT has since went to help out at Khan Academy - so it's really just Khan Academy now?


All times are GMT -7. The time now is 02:56 PM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
The contents of this forum are to be used ONLY by readers of the Learning From Data book by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin, and participants in the Learning From Data MOOC by Yaser S. Abu-Mostafa. No part of these contents is to be communicated or made accessible to ANY other person or entity.