LEC 8
Comp305 Part I. Artificial Neural Networks
Topic 3. Hebb’s Rules
1. Hebb’s Rules and the historical background
The
McColloch-Pitts
neuron made a base for a machine (network of
units) capable of
1. storing information and
2. producing logical and arithmetical operations on it
2. ANN learning rules
Definition:
ANN learning rule is
the rule how to adjust the
weights of connections to get
desirable output
.
Much work in Artificial Neural Networks focuses on the learning rules that define:
how to
change the weights of
connections
between neurons to better
adapt a network to serve some overall
function
.
3. Hebb’s Rule (1949)
a particular type of
use-dependent modification
of the connection strength of synapses
might
underlie
learning in the nervous system.
In another word:
“…
Cells that
fire together,
wire together
…”
•
The conditions that Hebb predicted would lead to changes in synaptic strength have now been found to cause the
long-term potentiation
in some neurons of
hippocampus
and other brain areas.
Hebb预测的会导致突触强度变化的条件现在已经被发现会导致海马体和其他大脑区域的一些神经元的长期增强。
The second equation emphasizes the
correlation nature of a Hebbian synapse
.
Sometimes the Hebb’s rule is referred to as
activity product rule
.
Hebb’s original learning rule referred
exclusively
to
excitatory synapses
第二个方程强调了赫比边突触的相关性。•有时Hebb规则被称为活动产品规则。•Hebb最初的学习规则只指兴奋性突触
