Turn down the lights
Turn up the lights

The Next Generation of Neural Networks

Google Tech Talks November, 29 2007 In the 1980's, new learning algorithms for neural networks promised to solve difficult classification tasks, like speech or object recognition, by learning many...  
 

More From: GoogleTechTalks

Next Generation All-IP Telecom Networks: Quality of Service Challenges and Is...56:30
42,078 views
Quantum Computing Day 1: Introduction to Quantum Computing56:27
53,110 views
Artificial intelligence and digital media56:43
21,922 views
Learning 3D Models from a Single Still Image58:38
17,734 views
Tangible Functional Programming56:23
24,756 views
Intelligence in Wikipedia51:03
18,536 views
Visual Perception with Deep Learning57:25
16,166 views
Aging of the Other Genome: A Decisive but Ambitious Solution1:02:26
24,514 views
A Possible Future of Software Development1:01:33
30,999 views
Cognitive Neuroscience of Mindfulness Meditation48:54
110,477 views
Large image databases and small codes for object recognition1:01:13
9,178 views
"Science and the taboo of psi" with Dean Radin1:34:57
74,183 views
Personal Growth Series: Karl Deisseroth on Cracking the Neural Code: Speaking...54:35
14,182 views
The Web That Wasn't59:35
38,354 views
Quantum Computing Day 2: Image Recognition with an Adiabatic Quantum Computer1:13:40
21,996 views
Wuala - a distributed file system48:33
56,862 views
Bayesian nonparametrics in document and language modeling1:03:49
5,622 views
No Time to Think58:08
66,763 views
Computational Neuroimaging1:02:40
3,939 views
The Neuroscience of Emotions1:02:10
29,837 views

QuickList(0)

Upgrade to Flash Player 10 for improved playback performance. Upgrade Now or get more info.
383 ratings
Sign in to rate
116,282 views
Want to add to Favorites? Sign In or Sign Up now!
Want to add to Playlists? Sign In or Sign Up now!
Want to flag a video? Sign In or Sign Up now!

Statistics & Data

Loading...

Video Responses (0)

This video has no Responses. Be the first to Post a Video Response.
Sign in to post a Comment

Text Comments (77)   Options

Loading...
Wulfcry (3 weeks ago) Show Hide
 0
Marked as spam
Although I love the way its scalable but hate the in between network calculation stuff one could redirect the network with the features if they sufficiently support the recognition part. But they suffer the brute force computing needed on the network its an if/if situation like the network to stochastic or confused. If only the features would be a more advanced data set.
eziooBand (3 weeks ago) Show Hide
 0
Marked as spam
wer guckt sich das denn 59:24 sekunden an will noch irgendwas sagen und naja
Wulfcry (3 weeks ago) Show Hide
 0
Marked as spam
Ja waarom niet het is een oude concept. En alhoewel het een novel idee wasje kunt het als herhaling stof bekijken. Dit concept is achterhaald heeft fouten en niet goed genoeg voor generaal gebruik zonder brute rekenkracht.
Probeer het toe te passen op een smartphone met een pxa270 dan merk je het.
Dirtfire (2 months ago)
Comment removed by author
shorty0802 (4 months ago) Show Hide
 0
Marked as spam
In 22:00, After presenting some number to the neural network. Shouldnt shouldnt it change the weights so the data matches more to one number, and when it runs backwards, why does it change it weights ?
GishForever (1 month ago) Show Hide
 0
Marked as spam
It uses back propagation to minimize the error.
kilianguntner (4 months ago) Show Hide
+1
Marked as spam
Interesting! :) //K
bluesrunthegame (5 months ago)
Comment removed by author
Comment(s) marked as spam Show
mallardvasey (6 months ago) Show Hide
+4
Marked as spam
Who cares?The only thing that matters is what he is saying, which is fascinating (if not entirely novel). If he wants to make a few gentle jokes to leaven things, more power to him.

Would you like to comment?

Join YouTube for a free account, or sign in if you are already a member.