Brain Dump

Correlation Matrix

Tags
adaptive-intelligence

Describes the [see page 21, correlation] between a collection of neurons.

For two neurons \( i,j \) the [see page 22, correlation] between them is defined as the average of the product of their output potential over a certain sample range: \[ C_{kj} = \langle v_k v_j \rangle = \frac{1}{P} \sum_^{P} v_k^{\mu} v_j^{\mu} \] The superscript \( \mu \) references the potential of the subscript neuron at time/sample \( \mu \).

We can define the [see page 22, correlation matrix] as a 2D matrix \( M[i][j] \) where each cell \( i,j \) contains the correlation between the neurons (\( i,j \)) \( C_{ij} \).

Links to this note