From: Dani Eder (firstname.lastname@example.org)
Date: Thu Jan 12 2006 - 08:14:47 MST
> On my machine, a 3GHz workstation, im able to run a
> feedforward network at
> about 150.000 operations /second WITH
> training(backprop) .. take training
> out of the equation and we may, lets shoot high,
> land on 1 million 'touched'
> neurons/second .. now from 10^6 -> 10^14 .. that's
> one hell of a big
You are losing a lot of orders of magnitude by
a neural network on a Von Neuman architecture. Taking
reverse, our brains, which are neural nets, can
a VN machine at about 1 op/sec (doing mathematical
We are much faster at things our brains are natively
like vision and speech recognition.
It would be a fairer comparison to compare the
capability of a workstation (~3 x 10^9/sec) to the
synapse firing rate. This implies that AI software
be designed to make efficient use of the CPU's
or alternatively, a 'neural net card' was installed to
accelerate that type of calculation. This would be
to graphics accelerators that are specialized for one
set of calculations.
A high end workstation, with four dual-core
have a theoretical rate of 24x10^9/sec, or a factor of
the synapse rate equivalent.
When I make these type of calculations, I use a factor
below and above the raw synapse rate to allow for
in how much data one synapse firing represents.
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
This archive was generated by hypermail 2.1.5 : Mon May 20 2013 - 04:01:03 MDT