DeepProbLog: Neural Probabilistic Logic Programming

TL;DR

DeepProbLog is a novel probabilistic logic programming language that integrates neural networks with logical reasoning, enabling end-to-end training and supporting symbolic and subsymbolic inference for program induction and learning.

Contribution

It introduces the first framework combining neural networks with probabilistic logic programming, allowing end-to-end training and versatile inference capabilities.

Findings

Supports symbolic and subsymbolic inference

Enables program induction and probabilistic reasoning

Demonstrates effective end-to-end training

Abstract

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 both symbolic and subsymbolic representations and inference, 1) program induction, 2) probabilistic (logic) programming, and 3) (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.