< Back to previous page

Project

Neural-symbolic artificial intelligence for real-world settings

Neural-symbolic artificial intelligence is the sub-field of AI that investigates the integration of deep learning and logic. This integration offers the best of both worlds: the learning capabilities of neural networks and the reasoning capabilities of logic. Although the field is moving at an amazing pace, it is not yet mature enough to be applied on real-world settings. One of the main reasons for this is that these neural-symbolic frameworks are not scalable. In the proposed work, we will resolve these issues by drawing inspiration from the related and more mature field of statistical-relational artificial intelligence.We will leverage the results from this field on probabilistic logic semantcs and approximate inference techniques and apply them on neural-symbolic frameworks. We will evaluate the new methods on large-scale datasets and the real-world setting of robotics.
Date:1 Jan 2022 →  31 Oct 2022
Keywords:Statistical relational artificial intell, Neural-symbolic intelligence, Deep learning
Disciplines:Knowledge representation and reasoning, Machine learning and decision making, Neural, evolutionary and fuzzy computation