Physics is the Best Teacher: Consistency Learning for Time-Invariant Operators of Chaotic Dynamics
cs.LG, nlin.CD
Submitted: 2026-10-02
Updated: 2026-10-02
Code: https://github.com/google/vizier4https:
Terminology
Sources
- FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators
- Chaos as an interpretable benchmark for forecasting and data-driven modelling
- Recurrent Neural Operators: Stable Long-Term PDE Prediction
- Physics-Informed Neural PDE Solvers via Spatio-Temporal MeanFlow
- Recovering Physical Dynamics from Discrete Observations via Intrinsic Differential Consistency
- Generalized Teacher Forcing for Learning Chaotic Dynamics
- Coarse Graining with Neural Operators for Simulating Chaotic Systems
- Identifying nonlinear dynamical systems via generative recurrent neural networks with applications to fMRI
- Neural Operator: Learning Maps Between Function Spaces
- On the difficulty of learning chaotic dynamics with RNNs
- Estimating fractal dimensions: a comparative review and open source implementations
- Data Driven Governing Equations Approximation Using Deep Neural Networks
- One Step Diffusion via Shortcut Models
- Deep-OSG: Deep Learning of Operators in Semigroup
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks