Neural-Symbolic Recursive Machine for Systematic Generalization
This addresses the challenge of human-like systematic generalization for AI systems that need to learn compositional rules from limited data and extrapolate to novel combinations.
The paper tackles the problem of systematic generalization in learning models by introducing the Neural-Symbolic Recursive Machine (NSR), which integrates neural perception, syntactic parsing, and semantic reasoning through a novel deduction-abduction algorithm. The results show NSR achieves superior performance over contemporary models on four benchmarks including SCAN, PCFG, HINT, and a compositional machine translation task.
Current learning models often struggle with human-like systematic generalization, particularly in learning compositional rules from limited data and extrapolating them to novel combinations. We introduce the Neural-Symbolic Recursive Machine (NSR), whose core is a Grounded Symbol System (GSS), allowing for the emergence of combinatorial syntax and semantics directly from training data. The NSR employs a modular design that integrates neural perception, syntactic parsing, and semantic reasoning. These components are synergistically trained through a novel deduction-abduction algorithm. Our findings demonstrate that NSR's design, imbued with the inductive biases of equivariance and compositionality, grants it the expressiveness to adeptly handle diverse sequence-to-sequence tasks and achieve unparalleled systematic generalization. We evaluate NSR's efficacy across four challenging benchmarks designed to probe systematic generalization capabilities: SCAN for semantic parsing, PCFG for string manipulation, HINT for arithmetic reasoning, and a compositional machine translation task. The results affirm NSR's superiority over contemporary neural and hybrid models in terms of generalization and transferability.