v1.4-beta.36

Additions

1. Multivariate student t distribution and random variable: MultivariateStudentsT and StudentsTRV
2. Students T Stochastic Process as outlined by Shah and Ghahramani
3. Added StationaryKernel into kernel API hierarchy.

v1.4-beta.35

Additions

1. Added variance control feature to CoupledSimulatedAnnealing

v1.4-beta.34

Improvements

1. Pretty print state of GlobalOptimizer using the companion object

v1.4-beta.33

Improvements

1. Optimised performance of CoupledSimulatedAnnealing

v1.4-beta.32

Improvements

1. Added variants of CoupledSimulatedAnnealing

1.4-beta.31

Additions

1. Sampling/simulation approaches to Stochastic Differential Equations and State Space Models i.e. Particle Filter. courtesy @amitkumarj441

Improvements

1. Faster computation of kernel matrices for CompositeCovariance objects.
2. Added MixedEffectRegularizer co-regionalization kernel for multi-output Gaussian Processes.

1.4-beta.30

Bugfixes

1. CoRegCauchyKernel: fix to hyper-parameter assignment

1.4-beta.29

Additions

1. Added MTGPRegressionModel for multi-task GP regression
2. Added MVGaussianScaler multivariate gaussian scaling of attributes
3. Added CoRegCauchyKernel coregionalization component for Cauchy Kernel

1.4-beta.28

Improvements

1. Changed :*: operator for kernels and random variables to :*

1.4-beta.27

Fix

1. Computation of kernel matrix between training and test points

Improvements

1. Computation of kernel matrix: speed