About: BCPy2000 provides a platform for rapid, flexible development of experimental BrainComputer Interface systems based on the BCI2000.org project. From the developer's point of view, the implementation [...] Changes:Bugfixes and tuneups, and an expanded set of (some more, some lessdocumented, optional tools)

About: The library is focused on implementation of propagation based approximate inference methods. Also implemented are a clique tree based exact inference, Gibbs sampling, and the mean field algorithm. Changes:Initial Announcement on mloss.org.

About: ELF provides many well implemented supervised learners for classification and regression tasks with an opportunity of ensemble learning. Changes:Initial Announcement on mloss.org.

About: dANN is an Artificial Intelligence and Artificial Genetics library targeted at employing conventional techniques as well as acting as a platform for research & development of novel techniques. As new techniques are developed and proven to be effective they will be integrated into the core library. It is currently written in Java, C++, and C#. However only the java version is currently in active development. If you want to obtain a version other than the java version you will need to get it directly from GIT. Changes:Please get the version in GIT only, the released version is old.

About: OpenKernel is a library for creating, combining, learning and using kernels for machine learning applications. Changes:Initial Announcement on mloss.org.

About: This software is an implementation of Hidden Markov Support Vector Machines (HMSVMs). Changes:Initial Announcement on mloss.org.

About: This software is designed for learning translation invariant kernels for classification with support vector machines. Changes:Initial Announcement on mloss.org.

About: GIDOC (Gimpbased Interactive transcription of old text DOCuments) is a computerassisted transcription prototype for handwritten text in old documents. It is a first attempt to provide integrated support for interactivepredictive page layout analysis, text line detection and handwritten text transcription. GIDOC is built on top of the wellknown GNU Image Manipulation Program (GIMP), and uses standard techniques and tools for handwritten text preprocessing and feature extraction, HMMbased image modelling, and language modelling. Changes:Updated version for mloss 2010

About: ALGLIB is an open source numerical analysis library distributed under GPL 2+. It implements both general numerical algorithms and machine learning algorithms. ALGLIB can be used from C#, C++, FreePascal, VBA and other languages. It is the only numerical analysis library which uses automatic translation to generate source code written in different programming languages with 100% identical functionality. Changes:

About: LIBSVM is an integrated software for support vector classification, (CSVC, nuSVC ), regression (epsilonSVR, nuSVR) and distribution estimation (oneclass SVM). It supports multiclass [...] Changes:Initial Announcement on mloss.org.

About: GPUML is a library that provides a C/C++ and MATLAB interface for speeding up the computation of the weighted kernel summation and kernel matrix construction on GPU. These computations occur commonly in several machine learning algorithms like kernel density estimation, kernel regression, kernel PCA, etc. Changes:Initial Announcement on mloss.org.

About: A fast implementation of several stochastic gradient descent learners for classification, ranking, and ROC area optimization, suitable for large, sparse data sets. Includes Pegasos SVM, SGDSVM, PassiveAggressive Perceptron, Perceptron with Margins, Logistic Regression, and ROMMA. Commandline utility and API libraries are provided. Changes:Initial Announcement on mloss.org.

About: A (randomized) coordinate descent procedure to minimize L1 regularized loss for classification and regression purposes. Changes:Fixed some I/O bugs. Lines that ended with whitespace were not read correctly in the previous version.

About: SeqAn is an open source C++ library of efficient algorithms and data structures for the analysis of sequences with the focus on biological data. Changes:

About: SHARK is a modular C++ library for the design and optimization of adaptive systems. It provides various machine learning and computational intelligence techniques. Changes:

About: Elefant is an open source software platform for the Machine Learning community licensed under the Mozilla Public License (MPL) and developed using Python, C, and C++. We aim to make it the platform [...] Changes:This release contains the Stream module as a first step in the direction of providing C++ library support. Stream aims to be a software framework for the implementation of large scale online learning algorithms. Large scale, in this context, should be understood as something that does not fit in the memory of a standard desktop computer. Added Bundle Methods for Regularized Risk Minimization (BMRM) allowing to choose from a list of loss functions and solvers (linear and quadratic). Added the following loss classes: BinaryClassificationLoss, HingeLoss, SquaredHingeLoss, ExponentialLoss, LogisticLoss, NoveltyLoss, LeastMeanSquareLoss, LeastAbsoluteDeviationLoss, QuantileRegressionLoss, EpsilonInsensitiveLoss, HuberRobustLoss, PoissonRegressionLoss, MultiClassLoss, WinnerTakesAllMultiClassLoss, ScaledSoftMarginMultiClassLoss, SoftmaxMultiClassLoss, MultivariateRegressionLoss Graphical User Interface provides now extensive documentation for each component explaining state variables and port descriptions. Changed saving and loading of experiments to XML (thereby avoiding storage of large input data structures). Unified automatic input checking via new static typing extending Python properties. Full support for recursive composition of larger components containing arbitrary statically typed state variables.

About: RLGlue allows agents, environments, and experiments written in Java, C/C++, Matlab, Python, and Lisp to inter operate, accelerating research by promoting software reuse in the community. Changes:RLGlue paper has been published in JMLR.

About: This package implements the “Online Random Forests” (ORF) algorithm of Saffari et al., ICCVOLCV 2009. This algorithm extends the offline Random Forests (RF) to learn from online training data samples. ORF is a multiclass classifier which is able to learn the classifier without 1vsall or 1vs1 binary decompositions. Changes:Initial Announcement on mloss.org.

About: EANT Without Structural Optimization is used to learn a policy in either complete or partially observable reinforcement learning domains of continuous state and action space. Changes:Initial Announcement on mloss.org.

About: Source code for EM approximate learning in the Latent Topic Hypertext Model. Changes:Initial Announcement on mloss.org.
