AMA i'm PhD Student of AI and working with Deep-learning
>>53430049
Are you me?
>>53430049
What's the predominant language you work with?
>>53430049
how's it feeling knowing that all your life's work only benefits the haves at the expense of the have-nots, while further widening the gap between them and doing nothing to solve any actual world problems?
>>53430049
Graduation subject?
>>53430049
Where do I get started.
Assume my highest math is precalc.
Also, are functional Lang's good for it?
I've hear lisp is good for it, but it's multi paradigm
I'm talking purely functional Lang's like Haskell.
>>53430103
He's right though
>>53430059
We all are one under our overlord
>>53430074
C++ and Python
>>53430086
good b8 btw
>>53430095
For masters : Using Deep-learning for recognizing food and estimate the calories by 3d reconstructing the food from photos
>PhD meme
>Deep Learning buzzword
>>53430049
How many peer-reviewed publications do you have?
![John-McCarthy-007[1].jpg](https://i.imgur.com/8mwTGIr.jpg "John-McCarthy-007[1].jpg")
>>53430126
>C++ and Python
Why?
>>53430105
This is good book to get to speed
>http://www.amazon.com/dp/0321204662/ref=cm_sw_su_dp
then coursera or udacity course on AI
>>53430130
yes it's just buzzword, but it's work wonders
>>53430135
3 , two of the is ISI
>>53430166
r u serious, that book cover doesn't look serious
also, this book is for beginner, right?
also, can you answer my questions on functional Lang's ?
>>53430159
C++ when i used to programming drones and robots.
but now is mostly Python for deep learning
>>53430166
>yes it's just buzzword, but it's work wonders
"it" is just comparing an item in a "list" against the other items. You're a fool
>>53430189
Get out frog meme poster
>>53430049
How do I motivate myself to push myself out of the shitty, stagnant NEET life I'm currently in and towards a doctorate?
How would you compare the degree/difficulty of work through your research compared to your previous degrees etc.?
>>53430204
gb2 /pol/ or /b/
>>53430181
The book is very good, and not too complicated, and it's good for beginners to get started, it's the second most used book to tech AI after this, with is very complicated
>http://www.amazon.com/Artificial-Intelligence-Modern-Approach-Edition/dp/0136042597/ref=zg_bs_491300_7
>>53430049
What is the main difference between the GA and EA?
How many layers do you need to call your network "deep"?
>>53430217
thanks senpai
>>53430217
Do you think their explanation is better than Murphy's? I only read the markov stuff in that book rly
>>53430221
>How many layers do you need to call your network "deep"?
as many as you can. Its all just a google search under the hood
>>53430252
Can I make an AI that recognizes good asses?
>>53430314
can you?
>>53430205
it's harder, but you should chooses someting you are interested in, and that will give the the push to learn more about it.
>>53430221
Do you mean Evolution strategies ?
if so :
ES are only a mutation operators, and it doesn't have Crossover, thus all genes are unique, and unlike GA mutation apply on all the population, and the replacement only happen it the offspring is better than the parent, otherwise we just clone the father into the new gen.
>How many layers do you need to call your network "deep"?
normally ANN have 1 to 3 layers, but deep nets can have 90 layers to work
How long does it take to learn enough to make something substantial.
>>53430049
What year did you graduated and from where?
>>53430332
is is possible?
>>53430383
Of course
>>53430314
>>53430383
yes we can, we just need your fap folder, and another folder with ugly asses, and you will have an ass classifier
>>53430342
it depend on you level of computer knowlage, but if you know fuzzbuzz, it will take you 6-12 months to make some great things, make a goal and move torword it
>>53430363
2014 - Malaya university
>>53430049
You can't be that good or you would have an AI do your PhD for you
>>53430407
So, essentially, AI is just patterns?
>good ass pattern
>bad ass pattern
>>53430126
>Using Deep-learning for recognizing food and estimate the calories by 3d reconstructing the food from photos
isn google working on exactly that atm?
can't wait to do shitpost learner or r9k beta learner
>>53430337
yep, evolutionary algorithms. had an oral test and switched GA and EA in that regard.
>>53430488
>evolutionary algorithm buzzword
>>53430506
is all you do shitpost?
>>53430217
>which is very complicated
my introduction AI class was based on that book, with sometimes 1:1 copies
good book,
with previous knowledge in general CS, a good start
>>53430437
big chunk of it, recognizing patterns.
>>53430457
yeah, but i think i can do better. *cry*
>>53430468
or you can make a bot the make shitpost and post it on reddit
>>53430488
Hope you get high marks
>>53430537
glad you liked it
>>53430049
What do you think are 5-10 'must-read' textbooks for AI/ML?
If I have a project that is a essentially a ML engine that predicts esports outcomes with good accuracy, is this too "toy-like" and "unprofessional" to put on my CV?
>>53430103
Yeah, bait or not, the guy is spot on.
>>53430588
>If I have a project that is a essentially a ML engine that predicts esports outcomes with good accuracy, is this too "toy-like" and "unprofessional" to put on my CV?
You should totally put that in your CV, but if you can make a version of it that work with normal sport, it will make bigger impact, but you should put that as it's.
>>53430619
You only put things on your CV if you have a following for them such as many blog posts about it, stars, etc. If you dont have this you have to either improve the project or network more or both
