Deep Learning for Symbolic Mathematics

Abstract

Neural networks have a reputation for being better at solving statistical or approximate problems than at performing calculations or working with symbolic data. In this paper, we show that they can be surprisingly good at more elaborated tasks in mathematics, such as symbolic integration and solving differential equations. We propose a syntax for representing mathematical problems, and methods for generating large datasets that can be used to train sequence-to-sequence models. We achieve results that outperform commercial Computer Algebra Systems such as Matlab or Mathematica.

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Sequence to Sequence Learning with Neural Networks
• Learning to Execute
• Neural GPUs Learn Algorithms
• Neural Arithmetic Logic Units
• Learning Continuous Semantic Representations of Symbolic Expressions
• Learning to Discover Efficient Mathematical Identities

Cited by in corpus (16)

• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence
• Generative Language Modeling for Automated Theorem Proving
• The Calabi-Yau Landscape: from Geometry, to Physics, to Machine-Learning
• Neural Symbolic Regression that Scales
• Training Verifiers to Solve Math Word Problems
• SymForce: Symbolic Computation and Code Generation for Robotics
• Graph Laplacians, Riemannian Manifolds and their Machine-Learning
• Computational model discovery with reinforcement learning
• Universes as Big Data
• Learning Potentials of Quantum Systems using Deep Neural Networks
• Compositional Generalization with Tree Stack Memory Units
• The DigitalTwin from an Artificial Intelligence Perspective
• Learning a performance metric of Buchberger's algorithm
• Towards Tractable Mathematical Reasoning: Challenges, Strategies, and Opportunities for Solving Math Word Problems
• Abstraction, Reasoning and Deep Learning: A Study of the "Look and Say" Sequence
• A neural network model of perception and reasoning