Sudoku Assistant – An AI Assistant Combining Machine Learning and Reasoning
We introduce The Sudoku Assistant, an AI assistant that integrates machine learning and reasoning to interpret, solve and explain pen-and-paper Sudokus scanned with a smartphone. The assistant demonstrates three increasingly important concepts in AI research: the integration of learning and reasoning, explainable AI and human-centered AI.
TorchicTab: Semantic Table Annotation
TorchicTab is a semantic table annotation system that automatically understands the content of a table and assigns semantic tags to its elements with high accuracy.
The Role of Counting in Probabilistic Reasoning
Do you know how to model count? We provide an introduction to model counting: what is it, why is it so important (hint: state-of-the-art probabilistic reasoning!), and what is the relation with knowledge compilation.
Lifted Reasoning for Combinatorial Counting
In this paper we develop lifted reasoning techniques for counting the number of valid configurations in combinatorics math word problems. Lifted reasoning exploits high-level symmetries and the interchangeability of objects to efficiently count the number of admissible assignments for a set of variables.
Real-time On-edge Acoustic Event Classification
Edge AI is the deployment of AI applications on embedded devices near the sensor, close to where the data is located, rather than centrally in a cloud computing facility or a data center.
Semantic web and knowledge graphs
A new research track, details available in 2022.
Latent Effects for Reusable Language Components
We have developed Latent Effects and Handlers, a new class of effects and handlers that support advanced control-flow mechanisms such as lazy evaluation, functions with effectful bodies or multistaging.
Clustering With User Feedback
This post is based on the following publications:
COBRAS: Interactive clustering with pairwise constraints. Van Craenendonck, T., Dumancic, S., Vanwolputte, E. & Blockeel, H. Intelligent Data Analysis, 2018. Tackling noise in active semi-supervised clustering.
Versatile Verification of Tree Ensembles using Veritas
This post is based on the following publications:
Laurens Devos, Wannes Meert, and Jesse Davis. “ Versatile Verification of Tree Ensembles.” To appear in the Proceedings of the 38th International Conference on Machine Learning.
User-Centric Logic-Based AI
An organisation’s greatest asset are often its employees, who posses a huge amount of knowledge and expertise about products and processes. We build AI systems that use this knowledge to help people make better and more efficient decisions.