Babel is a flexible toolkit for implementing and running agent-based experiments on emergent communication. The Babel system includes advanced modules for constructional language processing and learning, conceptualising and interpreting procedural semantic structures, and conducting multi-agent experiments in simulated environments or using physical robots.
Fluid Construction Grammar (FCG) is a special-purpose programming language designed for implementing computational construction grammars. It is mainly used as the language processing component in emergent communication experiments, for validating the consistency and preciseness of construction grammar models, and for enhancing the performance of language technology applications.
Parselmouth is a Python library for the Praat software. Parselmouth is unique in its aim to provide a complete and Pythonic interface to the internal Praat code.
A platform for the prediction and exploration of candidate disease-causing oligogenic variant combinations
DynaMine has been developed in Brussels thanks to a fruitful collaboration between different institutions.
VarCoPP is now also part of ORVAL, a newly published web-platform for the exploration of potential oligogenic disease signatures. ORVAL incorporates automated pre-filtering options for your VCF, variant combination pathogenicity predictions, predicted pathogenic gene networks and several PPI, pathway and prediction annotations.
State-of-the-art document layout segmentation technology. Layout analysis is a key step in many automatic document processing applications, such as analyzing invoices or pay slips, or labeling customer letters with topics. It recently won 2 prizes for best segmentation algorithm, beating Google’s Tesseract 4.
Multi-agents systems (MAS) are a core area of research in artificial intelligence. This toolbox is intended to be a visualization and simulation instrument in which we document state-of-the art methods and provide guidelines on when to use which approach.
Shepherd is a Django application that bridges state-of-the-art Reinforcement Learning agents (stable-baselines3) with a web API, for Reinforcement Learning on industrial machines, IoT, Matlab simulation, Javascript pages, etc. Shepherd allows several instances of a single agent to learn from each other (several thermostats performing the same task but in different homes can learn from each other transparently).