A simple English tutorial on artificial neural networks?

Question

I have Googled, StackOverflowed, that's it, and I can't find a tutorial that I can understand. I understand the concept of genetic algorithms and how to implement them (although I have not tried it), but I can not understand the concept of neural networks.

I dimly know how they work ... And about that. Can someone refer me to a textbook that could help someone who has not yet graduated from high school? Of course, I'm several years ahead of most people in my class, but I don't understand the summation (which I apparently need if I don't want simple binary output), and other things that I apparently should know.

Is there a simple, bare bonus for neural networks? After I learn the basics, I will move on to the more complex ones. Preferably, they would be in Java.

Thanks!

Answer 1

Summation just adds up a bunch of things. Thus,

Summation(1,2,3,4,5) = 1+2+3+4+5 = 15 

(note: it always adds: if you want to subtract, do the summation with negative numbers)

It was easy, wasn't it?;)

A vector is an ordered tuple that really just means a bunch of numbers in a specific order. Most often, physics describes the position, strength, speed, etc. .... this is really nothing special, just some ordered numbers, where the order is significant:

 v = <1,2,3> 

If we are talking about geometry, then this vector is a point in three-dimensional space, where the x coordinate is 1, the y coordinate is 2, and the z coordinate is 3 (see, which was also easy, right)?

In neural networks, a vector is usually a vector of inputs to a neuron, so it really is a list of numerical values. Summing a vector would be nothing more than summing up all the values ​​in the vector and resulting in a single number (which can be called a "scalar" value).

(it was thrown and simplified - I'm sure someone else will help me improve it;))

PS. Kudos to you for diving into this material at the high school level! :)

Answer 2

Well, there is this Wikipedia article Simple English, but I think you already know everything that it contains.

Answer 3

I had the same problem for a while. I am a high school student, so you are a little ahead of me. I got a vacation, and I used it to find out everything I could about the reverse distribution, and I did not find almost any resources that really help too much if you do not want to read so many calculi that you want to die. My advice is to write a perceptron first, which is a network with only input layers and output levels. It inspired me to write a message, so I hope that after half an hour of my posting there should be a tutorial at http://certioraomnia.blogspot.com/ . It may be a little late, as this question was asked three years ago, but it can help others later.