Machine learning a manifold Article Swipe

PDF

Related Concepts

Symmetry (geometry) Infinitesimal Representation (politics) Artificial neural network Scaling Simple (philosophy) Manifold (fluid mechanics) Mathematics Sigma Sampling (signal processing) Lie algebra Feature (linguistics) Algebra over a field Computer science Pure mathematics Artificial intelligence Mathematical analysis Filter (signal processing) Physics Geometry Linguistics Political science Engineering Quantum mechanics Law Philosophy Epistemology Mechanical engineering Politics Computer vision

Sean Craven , Djuna Croon , Daniel Cutting , R. Houtz ·

YOU? · · 2022 · Open Access · · DOI: https://doi.org/10.1103/physrevd.105.096030 · OA: W4226278505

We propose a simple method to identify a continuous Lie algebra symmetry in a dataset through regression by an artificial neural network. Our proposal takes advantage of the $ \mathcal{O}(ε^2)$ scaling of the output variable under infinitesimal symmetry transformations on the input variables. As symmetry transformations are generated post-training, the methodology does not rely on sampling of the full representation space or binning of the dataset, and the possibility of false identification is minimised. We demonstrate our method in the SU(3)-symmetric (non-) linear $Σ$ model.