| Author |
Message |
aaron
Joined: 25 Jun 2006
Posts: 7
07-18-06, 05:28 am |
 Post subject: How HTMs differ from Neural networks |
 |
|
I am curious as to how HTMs are different from the neural networks. are there substantial differences or merely old wine in new bottles.
can anyone, especially Numenta folks, offer some insights?
thanks
aaron |
|
| Back to top |
|
| Author |
Message |
FreeSynapse
Joined: 11 Jun 2006
Posts: 39
07-18-06, 10:52 pm |
| Post subject: |
|
|
Hello Aaron,
Well here is the academic answer about the similarites and differences of HTMs to existing AI technologies, particularly to Bayesian networks:
| Quote: | HTMs are similar to Bayesian networks; however, they differ from most Bayesian networks in the way that time, hierarchy, action, and attention are used.
An HTM can be considered a form of Bayesian network where the network consists of a collection of nodes arranged in a tree-shaped hierarchy. Each node in the hierarchy self-discovers a set of causes in its input through a process of finding common spatial patterns and then finding common temporal patterns. Unlike many Bayesian networks, HTMs are self-training, have a well-defined parent/child relationship between each node, inherently handle time-varying data, and afford mechanisms for covert attention. |
The above information is an excerpt from the HTM Concepts paper on Numenta's site. If you want the short answer, as I understand it, basic neural networks and bayesian networks just don't scale very well when it comes to forming very complex models, which is why this enhanced bayesian network is being developed. So I guess this is more like new wine in old bottles if I could call it that. |
|
| Back to top |
|
| Author |
Message |
Numenta-Phil
Joined: 28 Feb 2006
Posts: 62
Location: Menlo Park
07-19-06, 12:42 pm |
| Post subject: How HTMs differ from Neural networks |
|
|
Great question!
First of all, HTM's are a type of neural network. But in saying that, you should know that there are many different types of neural networks (single layer feedforward network, multi-layer network, recurrant, etc). 99% of these types of networks tend to emulate the neurons, yet don't have the overall infrastructure of the actual cortex.
Additionally, neural networks tend not to deal with temporal data very well, they ignore the hierarchy in the brain, and use a different set of learning algorithms that our implementation.
But, in a nutshell, HTMs are built according to biology.
Whereas neural networks ignore the structure and focus on the emulation of the neurons, HTMs tend to focus on the structure and ignores the emulation of the neurons.
I hope that clears things up.
_________________
Phillip B. Shoemaker
Director, Developer Services
Numenta, Inc. |
|
| Back to top |
|
 |
Page 1 of 1 |
All times are GMT - 8 Hours
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2002 phpBB Group
Please contact the board administrators if you have any questions regarding the OnIntelligence.org forums. |
| |
