AI Paper Atlas

Research map / Time series / Causal and generative time series

Tensor learning with orthogonal, Lorentz, and symplectic symmetries

Wilson G. Gregory, Josué Tonelli-Cueto, Nicholas F. Marshall, Andrew S. Lee, Soledad Villar

ICLR 2026 · Time series › Causal and generative time series · approach: Dynamical system reconstruction · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use equivariant model instead of MLP baselinestress-strain regression · TFENN dataset n=5,000test error: 1.586e-4 -> 4.057e-6Improved
use equivariant model instead of MLP baselinestress-strain regression · TFENN dataset n=40,000test error: 2.766e-5 -> 3.310e-6Improved
add rotation data augmentation to the MLPMLP baseline · TFENN dataset n=5,000test error: 1.586e-4 -> 2.020e-5Improved
use equivariant model instead of TFENNstress-strain regression · TFENN dataset n=20,000test error: 3.0e-5 -> 7.748e-7Improved
use equivariant model instead of MLP augmentedpath signature · O(d) pathssignature test error: 0.007 -> 0.002Improved
use equivariant model instead of MLP augmentedpath signature · Lorentz pathssignature test error: 0.186 -> 0.005Improved
use learned model instead of discrete signaturepath signature · Lorentz pathssignature test error: 1.489 -> 0.005Improved
use equivariant model instead of SoS with Random covariancesparse vector, Accept/Reject · Accept/Reject Randomsquared correlation: 0.610 -> 0.938Improved
use equivariant model instead of SoS with Identity covariancesparse vector, Accept/Reject · Accept/Reject Identitysquared correlation: 0.606 -> 0.190Hurt
use equivariant model instead of SoSsparse vector, Bernoulli-Gaussian · Bernoulli-Gaussian Randomsquared correlation: 0.962 -> 0.937Hurt
use equivariant model instead of SoSsparse vector, Bernoulli-Rademacher · Bernoulli-Rademacher Randomsquared correlation: 0.526 -> 0.957Improved
use equivariant model instead of MLP baselinesparse vector, Bernoulli-Rademacher · Bernoulli-Rademacher Randomsquared correlation: 0.923 -> 0.957Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy