Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

While trying to understand artificial neural networks, I came upon an equation for finding the net input of an artificial neuron. Can someone explain this to me and what it means? Here is the original article

$$y_i=f\left(\sum_j w_{ij}y_j\right)$$

Thanks in advance, ell.

share|improve this question
    
Are you talking about the "weighted sum" formula? –  Arturo Magidin Jan 31 '11 at 20:28
    
Are you referring to $$y_i=f\left(\sum_j w_{ij}y_j\right)$$ –  Isaac Jan 31 '11 at 20:30
    
Yes :) Is there any way I can put that on my post with my current reputation? –  Ell Jan 31 '11 at 20:39
add comment

2 Answers

up vote 3 down vote accepted

So hopefully you understand what the $\sum$ symbol means, it's just a summation across all of those indices. In this case we have our weight vectors

$w_1,w_2,\dots,w_n$

and our vector $y$. The neuron then is going to compute the new value of y_i, based on a dot product of w_i with y. Hence

$\sum_{j} w_{ij}y_j$

is just dot product of w_i and y, where both are n-dimensional vectors.

Then f is an activation function, it could be a preceptron or it could be a linear node. So it may output the value of the net, or it may output a 0 or a 1 based on the value of the summation. Or it could have some non-linear neuron it all depends on the network.

share|improve this answer
    
Thanks this helped a lot :) –  Ell Jan 31 '11 at 21:12
add comment

What you are looking at is a simplistic model of the neuron. In particular, the inputs are dendrites and the outputs are the synapse.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.