PED is a dialogue management system that uses a probabilistic nested belief model to choose dialogue strategies. The dialogue system designer need only supply a set of plan rules to PED as a dialogue grammar with preconditions. Using this grammar, PED constructs game trees (like the one below) to represent the outcomes of the dialogue, so that a dialogue strategy can be chosen for the current turn in the dialogue. PED automatically maintains a belief model by a belief revision process that uses the observed acts in the dialogue. The game tree is evaluated in the context of this belief model. PED is efficient because it uses probabilistic estimates of belief rather than a plain logical belief model.
Boosting is a meta-learning approach that aims at combining an ensemble of weak classifiers to form a strong classifier. Adaptive Boosting (Adaboost) implements this idea as a greedy search for a linear combination of classifiers by overweighting the examples that are misclassified by each classifier. icsiboost implements Adaboost over stumps (one-level decision trees) on discrete and continuous attributes (words and real values). This approach is one of the most efficient and simple to combine continuous and nominal values. This implementation is aimed at allowing training from millions of examples by hundreds of features in a reasonable amount of time/memory.
Python Web Graph Generator is a threaded Web graph (Power law random graph) generator. It can generate a synthetic Web graph of about one million nodes in a few minutes on a desktop machine. It supports both directed and undirected graphs. It implements a threaded variant of the RMAT algorithm. A little tweak can produce graphs representing social networks or community networks. It can also output connected components in a graph.
CellWriter is a grid-entry natural handwriting input panel. As you write characters into the cells, your writing is instantly recognized at the character level. When you press 'Enter' on the panel, the input you entered is sent to the currently focused application as if typed on the keyboard. Writer-dependent, CellWriter learns your handwriting for reliable recognition. Correcting preprocessor algorithms account for digitizer noise, differing stroke order, direction, and number of strokes. Unicode support enables you to write in any language.
dysii is a C++ library for distributed probabilistic inference and learning in large-scale dynamical systems. It provides methods such as the Kalman, unscented Kalman, and particle filters and smoothers, as well as useful classes such as common probability distributions and stochastic processes.
Urbi is a robotics software platform. It includes a C++/Java middleware API called UObject to interface components such as motors, cameras, and algorithms, and an innovative scripting language, urbiscript, with built-in support for parallel and event-based programming, used to write high-level behaviors and orchestrate the interactions between components. UObject components are built as shared libraries exposed as native objects within urbiscript, and either hot-plugged in a running Urbi engine, or started as a remote autonomous process communicating with the engine via the network. At any time, new urbiscript code can be sent to a running Urbi engine via a simple telnet, to introspect the state of components, modify existing code, or add new behaviors. Urbi is cross-platform and supports several robots (Gostai Jazz, Lego Mindstorms, Aldebaran Nao, Segway RMP, Spykee, Bioloid, etc.) and a simulator (Webots).
Logic Reasoner is a theorem prover for first-order logic with equality. The main objective leading the development of Logic Reasoner has been the creation of a flexible architecture: in particular, the program has been designed as a generic infrastructure for theorem proving, which forms the basis for a collection of specific proving techniques. These techniques can be easily combined or replaced to create configurations with different properties.