DeepAI AI Chat
Log In Sign Up

Learning Compositional Rules via Neural Program Synthesis

by   Maxwell I. Nye, et al.

Many aspects of human reasoning, including language, require learning rules from very little data. Humans can do this, often learning systematic rules from very few examples, and combining these rules to form compositional rule-based systems. Current neural architectures, on the other hand, often fail to generalize in a compositional manner, especially when evaluated in ways that vary systematically from training. In this work, we present a neuro-symbolic model which learns entire rule systems from a small set of examples. Instead of directly predicting outputs from inputs, we train our model to induce the explicit system of rules governing a set of previously seen examples, drawing upon techniques from the neural program synthesis literature. Our rule-synthesis approach outperforms neural meta-learning techniques in three domains: an artificial instruction-learning domain used to evaluate human learning, the SCAN challenge datasets, and learning rule-based translations of number words into integers for a wide range of human languages.


Learning Symbolic Rules for Reasoning in Quasi-Natural Language

Symbolic reasoning, rule-based symbol manipulation, is a hallmark of hum...

From Examples to Rules: Neural Guided Rule Synthesis for Information Extraction

While deep learning approaches to information extraction have had many s...

Data Vision: Learning to See Through Algorithmic Abstraction

Learning to see through data is central to contemporary forms of algorit...

PLANS: Robust Program Learning from Neurally Inferred Specifications

Recent years have seen the rise of statistical program learning based on...

Demonstration of CORNET: A System For Learning Spreadsheet Formatting Rules By Example

Data management and analysis tasks are often carried out using spreadshe...

CORNET: Learning Table Formatting Rules By Example

Spreadsheets are widely used for table manipulation and presentation. St...

RRULES: An improvement of the RULES rule-based classifier

RRULES is presented as an improvement and optimization over RULES, a sim...

Code Repositories


Code for "Learning Compositional Rules via Neural Program Synthesis"

view repo