Neural Probabilistic Logic Programming in DeepProbLog
This work proposes a novel framework for researchers in AI and machine learning, combining neural networks and probabilistic logic for enhanced modeling and reasoning, though it is foundational rather than incremental.
The authors tackled the problem of integrating deep learning with probabilistic logic programming by introducing DeepProbLog, a language that incorporates neural predicates, and demonstrated its ability to support symbolic and subsymbolic representations, program induction, and end-to-end learning from examples.
We introduce DeepProbLog, a neural probabilistic logic programming language that incorporates deep learning by means of neural predicates. We show how existing inference and learning techniques of the underlying probabilistic logic programming language ProbLog can be adapted for the new language. We theoretically and experimentally 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.