DeepProbLog: Neural Probabilistic Logic Programming.

• DOI: 10.48550/arxiv.1805.10872
• Publication Date: 2018-01-01
• Authors: Manhaeve, Robin; Dumancic, Sebastijan; Kimmig, Angelika; Demeester, Thomas; Raedt, Luc De
• Keywords: FOS: Computer and information sciences; Technology; Science & Technology; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; Computer Science; 01 natural sciences; Computer Science, Artificial Intelligence; 0105 earth and related environmental sciences

© 2018 Curran Associates Inc..All rights reserved. We introduce DeepProbLog, a probabilistic logic programming language that incorporates deep learning by means of neural predicates. We show how existing inference and learning techniques can be adapted for the new language. Our experiments demonstrate that DeepProbLog supports (i) both symbolic and subsymbolic representations and inference, (ii) program induction, (iii) probabilistic (logic) programming, and (iv) (deep) learning from examples. To the best of our knowledge, this work is the first to propose a framework where general-purpose neural networks and expressive probabilistic-logical modeling and reasoning are integrated in a way that exploits the full expressiveness and strengths of both worlds and can be trained end-to-end based on examples.<br/>status: Published online<br/>ispartof: Advances in Neural Information Processing Systems vol:31 pages:3753-3760<br/>ispartof: Thirty-second Conference on Neural Information Processing Systems location:Montreal, Canada date:2 Dec - 8 Dec 2018<br/>ispartof: pages:3753-3760<br/>