Hi Yuma,
It would be wonderful if you could put up your courses for others to
use. I have always been able to reason (and outreason) my professors
when it came to databases and programming concepts, but in maths, I feel
like a sheep being led by the horns. Not a good analogy, but you get my
drift. :-)
On 4/1/2016 1:51 AM, Yuma Decaux wrote:
Hi,
This is why I would like to start blogging about all of my math excursions from
blindness to understanding it all. The machine learning part was hard, yes, but
it's a process from fundamentals to higher mathematics which ML is clearly in.
If you don't know how vectors and matrices work, although they are easy to
express in matlab, you won't get much intuition into why one things does what.
I've been battling with my faculty heads for the past 3 years to get all their
material accessible, no simple task. But gradually with the people I meet, I am
able to assemble a pretty robust system for adapting these concepts to non
visual form.
Also, since I've taken the ML course and got a distinction for it, this means I
know the material inside out. My Tutor told me that as a sighted ML expert
doing a PHD, he thought it was a massive undertaking to do maths. So I'm not
sugar coating this, but just like guitars or other challenges, it's about the
hours put into it. If you don't put the hours, it won't work for you. The same
tutor had a few moments where he felt lacking in maths as I argued my way
mathematically as to why such and such does that and only that and why not
otherwise, etc. You acquire a mathematical sense of things, and it really works
for all computer science related courses since they all stem from the basis
language, maths.
I think I will try to put all my course notes up so you can understand each
concept. This will work very well for me too as it will challenge me to explain
it for visual impairment.
Cheers,
On 1/04/2016, at 6:00 AM, Lucas Radaelli <lucasradaelli@xxxxxxxxx> wrote:
I also tried to learn machine learning for a long time by myself.
it was very hard. finding accessible material, specially the math background
that is necessary was almost impossible.
Tried to follow some courses on coursera but did not have much success.
I think that the best way to approach this would be -- hire a bunch of people
to get a machine learning book and some math books and adapt them to us,
writing all the equations in LaTeX...
you know... this will require a lot of money unfortunately.
On 31/03/2016 11:47, Parham Doustdar wrote:
Hi,** To leave the list, click on the immediately-following link:-
I'm not finding any accessible way of learning mathematics online. I am trying
to learn machine learning, and I'm finding that I cannot understand the math
knowledge required for it, and when googling, I come up with charts and
formulae that cannot be read with screen readers.
I personally learn when the knowledge presented to me explains the 'why'. For
example, instead of just saying variance is calculated by deducting the mean
from each number in our data set, squaring the results, and then getting an
average, I would like to know why we are squaring the results there. As a
matter of fact, I spent about rhity minutes trying to figure out the answer to
this simple question on the net, and my answer was that this is to prevent our
results from canceling each other out, and to make them give a much more exact
number when the variance of our data is small.
Based on this method of learning, what resources do you recommend for learning
mathematics and statistics?
Thanks!
** To leave the list, click on the immediately-following link:-
** [mailto:program-l-request@xxxxxxxxxxxxx?subject=unsubscribe]
** If this link doesn't work then send a message to:
** program-l-request@xxxxxxxxxxxxx
** and in the Subject line type
** unsubscribe
** For other list commands such as vacation mode, click on the
** immediately-following link:-
** [mailto:program-l-request@xxxxxxxxxxxxx?subject=faq]
** or send a message, to
** program-l-request@xxxxxxxxxxxxx with the Subject:- faq
** [mailto:program-l-request@xxxxxxxxxxxxx?subject=unsubscribe]
** If this link doesn't work then send a message to:
** program-l-request@xxxxxxxxxxxxx
** and in the Subject line type
** unsubscribe
** For other list commands such as vacation mode, click on the
** immediately-following link:-
** [mailto:program-l-request@xxxxxxxxxxxxx?subject=faq]
** or send a message, to
** program-l-request@xxxxxxxxxxxxx with the Subject:- faq
