x
i
= a
i,1
s
1
+ · · · + a
i,k
s
k
+ · · · + a
i,n
s
n
weighted by the mixing weights a
i,k
. The same generative model can be
written in vectorial form as x = ¹
n
k=1
s
k
a
k
, where the observed random vector
x is represented by the basis vectors a
k
= (a
1,k
, · · · , a
m,k
)
T
. The basis vectors a
k
form the columns of the mixing matrix A = (a
1
, · · · , a
n
) and the generative
formula can be written as x = As, where s = (s
1
, · · · , s
n
)
T
. Given the model
and realizations (samples) x
1
, · · · , x
N
of the random vector x, the task is
to estimate both the mixing matrix A and the sources s. This is done by
adaptively calculating the vectors
Y
and setting up a cost function which
either maximizes the non-gaussianity of the calculated s
k
= (
Y
T
× x) or
minimizes the mutual information. In some cases, a prior knowledge of the
probability distributions of the sources can be used in the cost function. The
original sources s can be recovered by multiplying the observed signals x
with the inverse of the mixing matrix W = A
−1
, also known as the unmixing
matrix. Here it is assumed that the mixing matrix is square (n = m). If the
number of basis vectors is greater than the dimensionality of the observed
vectors, n > m, the task is overcomplete but is still solvable with the pseudo
inverse. In Linear noisy ICA model, with the added assumption of zeromean
and uncorrelated Gaussian noise n ~ N(0, diag(¹)), the ICA model takes the
form x = As + n. And in Non-linear ICA model, the mixing of the sources
does not need to be linear. Using a nonlinear mixing function f(·|θ) with
parameters θ, non-linear ICA model is x = f(s|θ) + n.
2.5.3 Non-negative Matrix Factorization
Non-negative matrix factorization (NMF) is a group of algorithms in
multivariate analysis and linear algebra where a matrix, X, is factorized
into (usually) two matrices, W and H: nmf(X) WH.
Factorization of matrices is generally non-unique, and a number
of different methods of doing so have been developed (e.g., principal
component analysis and singular value decomposition) by incorporating
different constraints; non-negative matrix factorization differs from these
methods in that it enforces the constraint that the factors W and H must be
non-negative, i.e., all elements must be equal to or greater than zero.
Let matrix V be the product of the matrices W and H such that:
WH = V
Matrix multiplication can be implemented as linear combinations of
column vectors in W with coeffi cients supplied by cell values in H. Each
column in V can be computed as follows:
v
i
=
1
N
ji j
j
Hw
=
∑
Mathematical Foundations 41