The KR&R group investigates modelling and representation of different forms of knowledge and reasoning, as found in a large variety of AI systems. We study both theoretical and applied aspects of knowledge representation and reasoning (KRR) formalisms, and try to apply these ideas to both scientific and real-world problems. From its early days on, our group has played an active role in the development of Semantic Web technologies. Also recently, we have broadened our focus to a wide-spectrum of problems in which KRR formalisms can help with: from neuro-symbolic AI to medicine and health care but also to collaborative multi-agent systems. These are also the elements of our research on Hybrid Intelligence (HI) which aims to combine human and machine intelligence. It is a long-term large scale project shared between 6 partner universities, headquartered by our group, and is funded by a 10 year Zwaartekracht grant from the Dutch Ministry of Education, Culture and Science.
Listed below are some of the publicly-available scientific and educational resources of the KR&R.
- burgerLinker
Command line tool for linking civil registries
- differentiable-fuzzy-logics
MNIST semi-supervised learning experiments using differentiable fuzzy logic
- juggl
*An interactive, stylable and expandable graph view for Obsidian. Juggl is designed as an advanced local graph view, where you can juggle * all your thoughts with ease.
- kgbench
A set of benchmark repositories for node classification on knowledge graphs.
- mkgfd
Discovering Context-Aware Constraints in Multimodal Knowledge Graphs
- mmlkg
A benchmark pipeline for shallow Multimodal Machine Learning on Knowledge Graphs
- mrgcn
Multimodal Relational Graph Convolution Network
- sameAs-webservice
*This Web service provides access to the largest collection of owl:sameAs statements and their equivalence classes after transitive * closure.
- storchastic
Stochastic Automatic Differentiation library for PyTorch.
- subMassive-webservice
SUBMASSIVE: Resolving Cyclic Relations in Very Large Knowledge Graphs
- torch-rgcn
A PyTorch implementation of the Relational Graph Convolutional Network (RGCN).
- mpqe
Answering queries over Knowledge Graphs with Graph Convolutional Networks @ ICLR 2020 Workshop on Graph Representation Learning
- blp
"Inductive Entity Representations from Text via Link Prediction" @ The Web Conference 2021
- cqd
Continuous Query Decomposition for Complex Query Answering in Incomplete Knowledge Graphs
- DLVU, deep learning at VU
- Knowledge & Data - Vrije Universiteit
Tutorials and Assignments for the Knowledge & Data course @ Vrije Universiteit
- Machine Learning @ VU University
Materials for the VU machine learning course
List updated on 2021-09-20