LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   The Final (http://book.caltech.edu/bookforum/forumdisplay.php?f=138)
-   -   Q21 (http://book.caltech.edu/bookforum/showthread.php?t=4115)

A M A 03-17-2013 10:04 AM

Q21
 
Q21 - Time to say thank-you isn't it ?


Thanks Dr. Yaser and all others. It was a good experience to take this course.


All the best.

Abdulbaset Ali

ripande 03-17-2013 11:27 AM

Re: Q21
 
Yes. Thanks to one and all in this forum who replied to my queries. It made a world of difference.

And special thanks to Prof Yaser, for the fabulous lectures and responding to all the queries from students on this forum so promptly

yaser 03-18-2013 01:55 AM

Re: Q21
 
Q21 :).

Thank you all.

Sendai 03-18-2013 04:06 PM

Re: Q21
 
Thanks to you, Professor, and everyone else who helped create this class. I really learned a lot.

ivankeller 03-18-2013 05:09 PM

Re: Q21
 
I want to thank a lot Pr. Abu-Mostafa and all the people who participate to make this excellent course possible.
I've learned so much in a different focus than another ML course I've taken before from Pr. Andrew Ng (Stanford UNiversity). In comparison, I can say that with your course I understood better the theoretical aspect of statistical learning. Ng's course was a bit more practical.
Ivan Keller.

heer2351 03-18-2013 05:59 PM

Re: Q21
 
Indeed thanks for this course I specifically learned a lot by making the homework.

Some feedback though, I have been doing quite a number of online courses over the last 2 years. What I miss in this course is some elaboration on the answers after the homework is closed.

In other courses the "evaluation" of the answers (also pointing to common mistakes) was very beneficial to the learning experience.

Haowen 03-18-2013 06:26 PM

Re: Q21
 
Quote:

Originally Posted by heer2351 (Post 10005)
What I miss in this course is some elaboration on the answers after the homework is closed.

I saw many instances of this kind of discussion on the forum for this class (see all the threads with *ANSWER* in the title). It's one of the advantages of a time-boxed class - everyone's doing the same homework at the same time and can help each other out.

jain.anand@tcs.com 03-18-2013 06:51 PM

Re: Q21
 
Thank you all for answering questions on the forum and participating so well. Without all your participation I would not have learned so much. Also, thanks a lot to Prof. Yasser and all the staff who made this course possible. This is the best machine learning course out there. Thank you !!!:bow:

melipone 03-18-2013 08:01 PM

Re: Q21
 
Yes, I really enjoyed the course. It had the right mix of theory and practice for computer scientists. The teacher was great and the participants in the forum were great! What I particularly appreciated was the flexibility to use our own programming language (although learning Octave for the vectorization constructs was very useful too).

ctallardc 03-19-2013 12:02 AM

Re: Q21
 
Thanks Dr. Yaser :bow: and everyone in the forum for his dedication and time to answer the questions.
This course was fantastic, far beyond my expectation.
The topics presented and the way professor could explain in simple term was super.
It has been one of the best course I have taken in my already long student life :D

I hope to use what I learned soon.

Thanks again



Carlos Tallard

ilya239 03-19-2013 03:03 PM

Re: Q21
 
Quote:

Originally Posted by ivankeller (Post 10003)
Ng's course was a bit more practical.

Actually I found this course more practical in the sense all solutions were built from scratch (in Ng's course you start with template code and fill in the key parts). It's useful to see for yourself that working classifiers can be built from scratch, and to go through some pitfalls of building them.

Thanks a lot Prof. Yaser for all you've put into this course. It's one of the best I've taken. The derivations were precise enough to help with understanding, without unnecessary details. The uniform "learning diagram" will be useful for approaching learning problems systematically. Building small learning systems yourself was good reassurance that this all actually works.

Thanks again, and all the best.

ilya

ajjcoppola 03-20-2013 10:25 AM

Re: Q21
 
Professor,
Just wanted to say thank you and your staff for a fantastic course!

A very "real-life" experience in that many times I either didn't know what I was doing or the time constraints made getting the programs done difficult.

In the end, I completed all the code, made many mistakes...fixed some...still have questions on others...but tried my best and learned a lot...thanks!

alan

sptripathi 06-09-2013 11:47 AM

Re: Q21
 
Thanks Prof Yaser and the team for the most wonderful course. Others have said enough, and I will only say - endless gratitude.
Looking forward to more courses from you. (wishful thinking)

yaser 06-09-2013 01:23 PM

Re: Q21
 
You are very kind.

PaulN 06-09-2013 02:01 PM

Re: Q21
 
Wonderful class indeed. Until the acknowledgments part, in the last lecture, I didn't realize how many people were involved. So thanks to everybody. Everything was top notch: the lectures of course, but also the production, the videography, the homework problems etc. And while other online classes can be interesting, useful or even inspiring, this is the first one I've taken which feels like the real thing, demanding, even at times exacting, but also fair. And the subject itself, Machine Learning, which I barely knew, is absolutely fascinating. It is a refreshing departure from other areas, and times, of AI which, while interesting in their own right, are (or were) very heuristic.

Learning is possible :)

yaser 06-09-2013 02:22 PM

Re: Q21
 
Quote:

Originally Posted by PaulN (Post 11060)
Wonderful class indeed. Until the acknowledgments part, in the last lecture, I didn't realize how many people were involved. So thanks to everybody. Everything was top notch: the lectures of course, but also the production, the videography, the homework problems etc. And while other online classes can be interesting, useful or even inspiring, this is the first one I've taken which feels like the real thing, demanding, even at times exacting, but also fair. And the subject itself, Machine Learning, which I barely knew, is absolutely fascinating. It is a refreshing departure from other areas, and times, of AI which, while interesting in their own right, are (or were) very heuristic.

It is wonderful to hear this assessment. Thank you.

marek 06-09-2013 03:51 PM

Re: Q21
 
Having just submitted the exam, I am incredibly impressed by what I experienced these past few months. I expected the challenge and rigor that comes with Caltech's reputation, but I was surprised by the quality of the presentation. It was accessible, intuitive, and painstakingly well organized. The global narrative tied together the individual lessons and helped reinforce each of the concepts. Ultimately, it felt like everything we were learning was useful and worth the effort. I thank you professor, and everyone that helped put this together, for this fantastic course.

Darcy Daugela 06-09-2013 06:03 PM

Thank-you
 
I just finished this class, and would like you to know just how much I appreciate this course. Due to my life circumstances I would not have been able to take this material in a classroom. When I started this class, I was not sure if what I would learn would be practical to my work.

Today I see so many opportunities to make use of Machine Learning. Your clear instruction combined with the hands on class assignments has helped both my understanding and confidence to start to make practical use of these concepts in the work place. This class has given me the opportunity to take that critical first step.

Thank-you.

yaser 06-09-2013 06:26 PM

Re: Q21
 
Quote:

Originally Posted by marek (Post 11064)
Having just submitted the exam, I am incredibly impressed by what I experienced these past few months. I expected the challenge and rigor that comes with Caltech's reputation, but I was surprised by the quality of the presentation. It was accessible, intuitive, and painstakingly well organized. The global narrative tied together the individual lessons and helped reinforce each of the concepts. Ultimately, it felt like everything we were learning was useful and worth the effort. I thank you professor, and everyone that helped put this together, for this fantastic course.

I appreciate your kind words, and the detailed positive feedback.

yaser 06-09-2013 06:28 PM

Re: Thank-you
 
Quote:

Originally Posted by Darcy Daugela (Post 11066)
I just finished this class, and would like you to know just how much I appreciate this course. Due to my life circumstances I would not have been able to take this material in a classroom. When I started this class, I was not sure if what I would learn would be practical to my work.

Today I see so many opportunities to make use of Machine Learning. Your clear instruction combined with the hands on class assignments has helped both my understanding and confidence to start to make practical use of these concepts in the work place. This class has given me the opportunity to take that critical first step.

Thank-you.

I am happy that the course has had this positive impact for you. My best wishes as you take further steps to learn more of this fascinating and dynamic subject.

msram 06-11-2013 10:46 AM

Re: Q21
 
Thank you so much Professor. After doing this course, my confidence in machine learning has grown significantly. :bow:

Ram

yaser 06-11-2013 01:13 PM

Re: Q21
 
Quote:

Originally Posted by msram (Post 11114)
Thank you so much Professor. After doing this course, my confidence in machine learning has grown significantly. :bow:

Ram

Thanks, Ram. Glad to hear.

OlivierB 06-12-2013 02:20 PM

Re: Q21
 
Thank you VERY MUCH for this great course ("the real course, not a watered down version") !! Indeed, it proved more work than I anticipated, but I feel what I learnt will stay. Clearly, it was way better than just a collection of recipes. I found the approach the very right balance between theory and practice. I appreciated the learning path your have chosen, the rigor in each reasoning step (going as far as possible with intuition and then calling out clearly identified mathematical methods), and hey, they were entertaining !! I found the book a very useful reference too. All in all, close to perfect, in my view. Here is the grade I would give you: 9.99/10 :D

Having been able to follow this superb course 9 time zones away from Caltech, while I have quite a busy life is a marvelous feast of technology !

I understand this is the final session of the course. I was just about to strongly recommend it... May I ask why ? Too much time to conduct ? Are you busy on other projects ? Or are you thinking of changing the format ?

Anyway, I sincerely thank you and your team !
Olivier

yaser 06-12-2013 02:45 PM

Re: Q21
 
Thank yoiu for your kind words.

Quote:

Originally Posted by OlivierB (Post 11139)
I understand this is the final session of the course. I was just about to strongly recommend it... May I ask why ? Too much time to conduct ? Are you busy on other projects ? Or are you thinking of changing the format ?

The course will continue in a self-paced mode, both on iTunes U and on the course site, so please feel free to let others know. This forum and the Machine Learning Video Library will also continue.

The reason I am not offering more regular sessions is the significant time demand they pose. I do not wish to offer a session without attending to it properly.

robbinsleep 03-12-2016 06:54 AM

Re: Q21
 
Looking forward to use what I have learned. Thanks again

Robbin


All times are GMT -7. The time now is 06:49 AM.

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.