TensorLog: A Differentiable Deductive Database

Abstract

Large knowledge bases (KBs) are useful in many tasks, but it is unclear how to integrate this sort of knowledge into "deep" gradient-based learning systems. To address this problem, we describe a probabilistic deductive database, called TensorLog, in which reasoning uses a differentiable process. In TensorLog, each clause in a logical theory is first converted into certain type of factor graph. Then, for each type of query to the factor graph, the message-passing steps required to perform belief propagation (BP) are "unrolled" into a function, which is differentiable. We show that these functions can be composed recursively to perform inference in non-trivial logical theories containing multiple interrelated clauses and predicates. Both compilation and inference in TensorLog are efficient: compilation is linear in theory size and proof depth, and inference is linear in database size and the number of message-passing steps used in BP. We also present experimental results with TensorLog and discuss its relationship to other first-order probabilistic logics.

Cited by in corpus (23)

• Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning
• Differentiable Learning of Logical Rules for Knowledge Base Reasoning
• End-to-End Differentiable Proving
• Logical Neural Networks
• Inductive Relation Prediction by Subgraph Reasoning
• Towards End-to-End Reinforcement Learning of Dialogue Agents for Information Access
• Beyond Graph Neural Networks with Lifted Relational Neural Networks
• Logical Rule Induction and Theory Learning Using Neural Theorem Proving
• Dynamically Pruned Message Passing Networks for Large-Scale Knowledge Graph Reasoning
• Constraint-Based Visual Generation
• DeepLogic: Towards End-to-End Differentiable Logical Reasoning
• Learning First-Order Rules with Relational Path Contrast for Inductive Relation Reasoning
• Modeling Attention Flow on Graphs
• Neural, Symbolic and Neural-Symbolic Reasoning on Knowledge Graphs
• Enhancing Linear Algebraic Computation of Logic Programs Using Sparse Representation
• NLProlog: Reasoning with Weak Unification for Question Answering in Natural Language
• Stacked Structure Learning for Lifted Relational Neural Networks
• Neural Consciousness Flow
• Differentiable Satisfiability and Differentiable Answer Set Programming for Sampling-Based Multi-Model Optimization
• pix2rule: End-to-end Neuro-symbolic Rule Learning
• Learning Invariants through Soft Unification
• Neural Networks Enhancement with Logical Knowledge
• A Statistical Relational Approach to Learning Distance-based GCNs