.. _functional_chaos:

Functional Chaos Expansion
--------------------------

| Accounting for the joint probability density function (PDF)
  :math:`f_{\underline{X}}(\underline{x})` of the input random vector
  :math:`\underline{X}`, one seeks the joint PDF of the model response
  :math:`\underline{Y} = h(\underline{X})`. This may be achieved using
  Monte Carlo (MC) simulation, i.e. by evaluating the model :math:`h`
  at a large number of realizations :math:`\underline{x}^{(i)}` of
  :math:`\underline{X}` and then by estimating the empirical
  distribution of the corresponding sample of model output
  :math:`h(\underline{x}^{(i)})`. However it is well-known that the MC
  method requires a large number of model evaluations, i.e. a great
  computational cost, in order to obtain accurate results.
| In fact, when using MC simulation, each model run is performed
  independently. Thus, whereas it is expected that
  :math:`h(\underline{x}^{(i)}) \approx h(\underline{x}^{(j)})` if
  :math:`\underline{x}^{(i)} \approx \underline{x}^{(j)}`, the model is
  evaluated twice without accounting for this information. In other
  words, the functional dependence between :math:`\underline{X}` and
  :math:`\underline{Y}` is lost.
| A possible solution to overcome this problem and thereby to reduce the
  computational cost of MC simulation is to represent the random
  response :math:`\underline{Y}` in a suitable functional space, such as
  the Hilbert space :math:`L^2` of square-integrable functions with
  respect to the PDF :math:`f_{\underline{X}}(\underline{x})`.
  Precisely, an expansion of the model response onto an orthonormal
  basis of :math:`L^2` is of interest.

| The principles of the building of a (infinite numerable) basis of this
  space, i.e. the PC basis, are described in the sequel.
| **Principle of the functional chaos expansion**
| Consider a model :math:`h` depending on a set of *random* variables
  :math:`\underline{X} = (X_1,\dots,X_{n_X})^{\textsf{T}}`. We call
  functional chaos expansion the class of spectral methods which gathers
  all types of response surface that can be seen as a projection of the
  physical model in an orthonormal basis. This class of methods uses the
  Hilbertian space (square-integrable space: :math:`L^2`) to construct
  the response surface.
| Assuming that the physical model has a finite second order measure
  (i.e. :math:`E\left( \|h(\underline{X})\|^2\right)< + \infty`), it may
  be uniquely represented as a converging series onto an orthonormal
  basis as follows:

  .. math::

      h(\underline{x})= \sum_{i=0}^{+\infty}  \underline{y}_{i}\Phi_{i}(\underline{x}).

  where the
  :math:`\underline{y}_{i} = (y_{i,1},\dots,y_{i,n_Y})^{\textsf{T}}`\ ’s
  are deterministic vectors that fully characterize the random vector
  :math:`\underline{Y}`, and the :math:`\Phi_{i}`\ ’s are given basis
  functions (e.g. orthonormal polynomials, wavelets).

| The orthonormality property of the functional chaos basis reads:

  .. math::

     \langle \Phi_{i},\Phi_{j}\rangle = \int_{D}\Phi_{i}(\underline{x}) \Phi_{j}(\underline{x})~f_{\underline{X}}(\underline{x}) d \underline{x} = \delta_{i,j}.

  where :math:`\delta_{i,j} =1` if :math:`i=j` and 0 otherwise. The
  metamodel :math:`\widehat{h}(\underline{x})` is represented by a
  *finite* subset of coefficients
  :math:`\{y_{i}, i \in \cA \subset (N)\}` in a *truncated* basis
  :math:`\{\Phi_{i}, i \in \cA \subset (N)\}` as follows:

  .. math::

      \widehat{h}(\underline{x})= \sum_{i \in \cA \subset N}  y_{i}\Phi_{i}(\underline{x})

As an example of this type of expansion, one can mention responses by
wavelet expansion, polynomial chaos expansion, etc.


.. topic:: API:

    - See :class:`~openturns.FunctionalChaosAlgorithm`


.. topic:: Examples:

    - See :doc:`/auto_meta_modeling/polynomial_chaos_metamodel/plot_functional_chaos`


.. topic:: References:

    - [soizeghanem2004]_