What math do I need to know/study to master AI?

Fagonomics

>>936223397 (OP)
tensors, presumably.

>>936223397 (OP)
what a pussy

>>936223397 (OP)
tensor arithmetic

>>936223397 (OP)
You don't need to know anything to master AI, and certainly not math. Trust Mr. Jimmy. Listen to Mr. Jimmy. Mr. Jimmy is your friend. Mr. Jimmy knows what's best for you. Mr. Jimmy wants what's best for you.

>>936223397 (OP)
higher math, college stuff. if you're lucky, high school AP math covers it.

>>936224215
>>936225857
no. wrong.
they call them "tensors" but they are nothing but high-dimensional arrays.
"tensors" have nothing to do with what real tensors are used for in physics or math.

>>936223397 (OP)
most of AI these days is equivalent to biology and anatomy. some fundamentals, but mostly lots of "structures" that have proven to be advantageous.
if you can study any subject *other than* math, you can do well with AI.

>>936227866
>In machine learning, the term tensor informally refers to two different concepts (i) a way of organizing data and (ii) a multilinear (tensor) transformation. Data may be organized in a multidimensional array (M-way array), informally referred to as a "data tensor"; however, in the strict mathematical sense, a tensor is a multilinear mapping over a set of domain vector spaces to a range vector space. Observations, such as images, movies, volumes, sounds, and relationships among words and concepts, stored in an M-way array ("data tensor"), may be analyzed either by artificial neural networks or tensor methods.

>>936227927
you asked the chat bot, didn't you

>>936224215
>>936225857
Not really.

In rough order of importance
> linear algebra, multivariate calculus, combinatorics, taylor expansions
covers like 95% of ML models

>>936228011
no that's literally the quote from wikipedia /b/ro

>>936228018
>linear
buildup to tensors
>multi variable calc
buildup to tensors
>combinatorics
general stas
>taylor expansions
build up to actual tensor arithemtic
if you learn tensor arithmetic you learn all of that stuff anon

>>936228223
wrong page though.
https://en.wikipedia.org/wiki/Tensor
that's a real tensor.
"tensor" in neural networks is nothing but a multidimensional array they do some simple operations on.

>>936223397 (OP)
calculus to start with, minimum, then work up from there

Start with the book “Why Machines Learn” by Anil Ananthaswc. It’ll give you an overview and you can go from there.

>>936227866
>"tensors" have nothing to do with what real tensors are used for in physics or math.
ummmmm not really sweetie
tensors in muhhhhhh real physics are high-dimensional arrays perhaps with some transformation properties
they have a lot to do with high-dimensional arrays

>>936223397 (OP)
There is a lot of math if you want to master AI similar to mastering math (masters degree). To name a few for image generation:

>Convolutional Operations
>Residual Connections
>Element-wise addition
>Attention Mechanisms like Scaled dot-product attention and Attention with external input
>Group Normalisation
>Sinusoidal Time Embeddings

All of these are for the neural network that learns how to turn one image into another (which is what img gen is). You can find what you need exactly by reading the paper of any AI model or algorithm.

ai is still in infant stages.
the way the infant sees what it wants, and assumes that knowledge has declared it being obtainable/inherited by default. however, the price of knowledge, of true math, is on an economical table outside the scope of ai or even human systems. You will need to know how 1+2 = 3 where 1 is human intelligence, 2 is the shared relation of an index and distribution table (or ai as it is)---in all its developments this is limited drastically by an impossible demand in knowledge retrieval and computational benchmarks. This is where we are at. But merging Human and AI is an extinction event the world must be prepared for. Also, its complicated enough to require quantum mechanics/computation/physics etc rolled up into one package. 2+1 = a +b +c. You will basically have to see beyond what it means to be human, or machine, and to understand what it means to be God. The world is close, but not ready to handle itself like the Matrix itself. For that, you will need faith.

>>936223397 (OP)
aids math

Let's spice things up a little, shall we? emaquiie