AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularity

Abstract

We present an improved method for symbolic regression that seeks to fit data to formulas that are Pareto-optimal, in the sense of having the best accuracy for a given complexity. It improves on the previous state-of-the-art by typically being orders of magnitude more robust toward noise and bad data, and also by discovering many formulas that stumped previous methods. We develop a method for discovering generalized symmetries (arbitrary modularity in the computational graph of a formula) from gradient properties of a neural network fit. We use normalizing flows to generalize our symbolic regression method to probability distributions from which we only have samples, and employ statistical hypothesis testing to accelerate robust brute-force search.

17 pages, 6 figs, replaced to match accepted NeurIPS version

References in corpus (10)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Variational Inference with Normalizing Flows
• Normalizing Flows: An Introduction and Review of Current Methods
• Density estimation using Real NVP
• Mutual Information, Neural Networks and the Renormalization Group
• Toward an AI Physicist for Unsupervised Learning
• Increasing the Interpretability of Recurrent Neural Networks Using Hidden Markov Models
• Symbolic Pregression: Discovering Physical Laws from Distorted Video
• A Theory of Universal Artificial Intelligence based on Algorithmic Complexity
• Comment on the article "Distilling free-form natural laws from experimental data"

Cited by in corpus (23)

• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• Contemporary Symbolic Regression Methods and their Relative Performance
• AI Poincaré: Machine Learning Conservation Laws from Trajectories
• Lattice Thermal Conductivity Prediction using Symbolic Regression and Machine Learning
• Machine-learning hidden symmetries
• Symbolic Pregression: Discovering Physical Laws from Distorted Video
• Precision Machine Learning
• Probabilistic Grammars for Equation Discovery
• Data-Driven Equation Discovery of a Cloud Cover Parameterization
• Neural Symbolic Regression that Scales
• Data-driven, multi-moment fluid modeling of Landau damping
• Iterative Symbolic Regression for Learning Transport Equations
• MMSR: Symbolic Regression is a Multi-Modal Information Fusion Task
• PGD-based advanced nonlinear multiparametric regressions for constructing metamodels at the scarce-data limit
• Information Fusion via Symbolic Regression: A Tutorial in the Context of Human Health
• From Kepler to Newton: Explainable AI for Science
• cp3-bench: A tool for benchmarking symbolic regression algorithms tested with cosmology
• Analysis of strong coupling constant with machine learning and its application
• Deep learning complete intersection Calabi-Yau manifolds
• Logic Guided Genetic Algorithms
• Solution to a class of multistate Landau-Zener model beyond integrability conditions
• Learning Physical Concepts in Cyber-Physical Systems: A Case Study
• Physics-Augmented Learning: A New Paradigm Beyond Physics-Informed Learning