Specifying Derivatives (Mathematica Tutorial)

The function FindRoot has a Jacobian option; the functions FindMinimum, FindMaximum, and FindFit have a Gradient option; and the "Newton" method has a method option Hessian. ...

Termination Conditions (Mathematica Tutorial)

Mathematically, sufficient conditions for a local minimum of a smooth function are quite straightforward: x^* is a local minimum if ∇f(x^*)=0 and the Hessian ∇^2f(x^*) is ...

Test Problems (Mathematica Tutorial)

All the test problems presented in [MGH81] have been coded into Mathematica in the Optimization`UnconstrainedProblems` package. A data structure is used so that the problems ...

Trust Region Methods (Mathematica Tutorial)

A trust region method has a region around the current search point, where the quadratic model for "local minimization" is "trusted" to be correct and steps are chosen to stay ...

Variables and Starting Conditions (Mathematica Tutorial)

All the functions FindMinimum, FindMaximum, and FindRoot take variable specifications of the same form. The function FindFit uses the same form for its parameter ...

CUDALink (CUDALink Guide)

CUDALink allows Mathematica to use the CUDA parallel computing architecture on Graphical Processing Units (GPUs). It contains functions that use CUDA-enabled GPUs to boost ...

CUDAArgMaxList (CUDALink Symbol)

CUDAArgMaxList[list] gives the index of the maximum element in list.

CUDAArgMinList (CUDALink Symbol)

CUDAArgMinList[list] gives the index of the minimum element in list.

CUDABoxFilter (CUDALink Symbol)

CUDABoxFilter[img, r] gives the box filter of img with radius r.CUDABoxFilter[list, r] gives the box filter of list with radius r.CUDABoxFilter[mem, r] gives the box filter ...

CUDACCompilers (CUDALink Symbol)

CUDACCompilers[] gives a list of C compilers detected on system and supported by NVCCCompiler.