We present an implementation of a probabilistic first-order logic called TensorLog, in which classes of logical queries are compiled into differentiable functions in a neural-network infrastructure such as Tensorflow or Theano. This leads to a close integration of probabilistic logical reasoning with deep-learning infrastructure: in particular, it enables high-performance deep learning frameworks to be used for tuning the parameters of a probabilistic logic. The integration with these frameworks enables use of GPU-based parallel processors for inference and learning, making TensorLog the first highly parallellizable probabilistic logic. Experimental results show that TensorLog scales to problems involving hundreds of thousands of knowledge-base triples and tens of thousands of examples.

In predicate invention (PI), new predicates are introduced into a logical theory, usually by rewriting a group of closely-related rules to use a common invented predicate as a "subroutine". PI is difficult, since a poorly-chosen invented predicate may lead to error cascades. Here we suggest a "soft" version of predicate invention: instead of explicitly creating new predicates, we implicitly group closely-related rules by using structured sparsity to regularize their parameters together. We show that soft PI, unlike hard PI, consistently improves over previous strong baselines for structure-learning on two large-scale tasks.

A key challenge in statistical relational learning is to develop a semantically rich formalism that supports efficient probabilistic reasoning using large collections of extracted information. This paper presents a new, scalable probabilistic logic called ProPPR, which further extends stochastic logic programs (SLP) to a framework that enables efficient learning and inference on graphs: using an abductive second-order probabilistic logic, we show that first-order theories can be automatically generated via parameter learning; that in parameter learning, weight learning can be performed using parallel stochastic gradient descent with a supervised personalized PageRank algorithm; and that most importantly, queries can be approximately grounded with a small graph, and inference is independent of the size of the database.