My Math Forum  

Go Back   My Math Forum > Science Forums > Computer Science

Computer Science Computer Science Forum


Reply
 
LinkBack Thread Tools Display Modes
November 28th, 2012, 11:37 AM   #1
Newbie
 
Joined: Nov 2012

Posts: 1
Thanks: 0

Neural networks - backpropagation

Hi,

I am trying to learn neural networks and I am reading a book called: Machine Learning:
An Algorithmic Perspective. This book can be found from google books. Hopefully this link works: (http://books.google.com/books?id=n66O8a4SWGEC&hl=fi)

I am not able to follow what happens in formula 3.48 on page 90. I can not understand why the h_k is equivalent with only one activation function instead of sum of activation functions multiplied with weights.

Can somebody explain origin of formula 3.48 in more details?

Thanks
sampie is offline  
 
Reply

  My Math Forum > Science Forums > Computer Science

Tags
backpropagation, networks, neural



Thread Tools
Display Modes


Similar Threads
Thread Thread Starter Forum Replies Last Post
Represent continuum betw summary and detail of timed neural BenFRayfield Number Theory 0 January 19th, 2014 07:02 PM
Bayesian Networks mhhojati Advanced Statistics 0 January 6th, 2014 07:11 AM
Bayesian Networks David A Advanced Statistics 0 November 1st, 2013 08:03 PM
We understand sizes changing because Neural Doppler effect BenFRayfield Complex Analysis 1 August 1st, 2013 04:27 PM
Neural networks Spook Computer Science 1 August 18th, 2012 06:11 AM





Copyright © 2018 My Math Forum. All rights reserved.