Shark Machine Learning Library

Summary

SHARK is a modular C++ library for the design and optimization of adaptive systems. It provides methods for linear and nonlinear optimization, in particular evolutionary and gradient-based algorithms, kernel-based learning algorithms and neural networks, and various other machine learning techniques. SHARK serves as a toolbox to support real world applications as well as research in different domains of computational intelligence and machine learning. The sources are compatible with the following platforms: Windows, Solaris, MacOS X, and Linux.

Pre-built packages for Debian/Ubuntu and Windows as well as a source package are available in the following locations:

Features

The Shark library comprises four main modules:

ReClaM: A machine learning framework for regression and classification tasks, including neural networks and kernel methods.
EALib: An evolutionary computation framework for solving discrete and continous optimization problems.
MOO-EALib: An extension of the EALib for multi-objective optimization.
Fuzzy: A framework for fuzzy logic and control.

Additionally, multiple support modules are included with Shark:

Mixture: Representation and optimization of mixture density models.
Array: Multi-dimensional arrays of variable size.
LinAlg: Linear algebra algorithms, e.g., singular value decomposition and eigen value decomposition.
FileUtil: I/O and management of configuration files.

Support

Multiple support channels are available for the Shark library. If you are a new user, please refer to the section Getting Started and to the section FAQ. If you are interested in the documentation of the library, refer to the respective module you are interested in. If there are any questions left, do not hesitate to contact us via email or to post to the Shark user mailing-list available at https://lists.sourceforge.net/lists/listinfo/shark-project-user.

Paper Describing the Library

Christian Igel, Verena Heidrich-Meisner, and Tobias Glasmachers. Shark. Journal of Machine Learning Research 9, pp. 993-996, 2008

@Article{shark08,
  author = {Christian Igel and Tobias Glasmachers and Verena Heidrich-Meisner},
  title = {Shark},
  journal = {Journal of Machine Learning Research},
  year = {2008},
  volume = {9},
  pages = {993-996},
}

Acknowledgements

The development of the Shark library is supported by the following institutions and companies: