mlsanityData to 14 Oct 2026

Generalizing to New Physical Systems via Context-Informed Dynamics Model

arXiv 2202.01889 ↗First posted 1 Feb 2022cs.LG cs.AI stat.MLcited by 50

Not on the map. Why?

Abstract (arXiv)

Data-driven approaches to modeling physical systems fail to generalize to unseen systems that share the same general dynamics with the learning domain, but correspond to different physical contexts. We propose a new framework for this key problem, context-informed dynamics adaptation (CoDA), which takes into account the distributional shift across systems for fast and efficient adaptation to new dynamics. CoDA leverages multiple environments, each associated to a different dynamic, and learns to condition the dynamics model on contextual parameters, specific to each environment. The conditioning is performed via a hypernetwork, learned jointly with a context vector from observed data. The proposed formulation constrains the search hypothesis space to foster fast adaptation and better generalization across environments. We theoretically motivate our approach and show state-of-the-art generalization results on a set of nonlinear dynamics, representative of a variety of application domains. We also show, on these systems, that new system parameters can be inferred from context vectors with minimal supervision. Code is available at https://github.com/yuan-yin/CoDA .

Key facts. Cited by 50 papers in the map (data through 14 Oct 2026). Builds on 43 of its 65 references that are in the map. First posted on arXiv 1 Feb 2022.

Builds on

43 of 65 references.

References · 43

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  3. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  5. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  6. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  7. 2019Invariant Risk Minimizationcited by 1,854
  8. 2017Visualizing the Loss Landscape of Neural Netscited by 1,130
  9. 2016HyperNetworkscited by 1,027
  10. 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 803
  11. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 766
  12. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
31 more
  1. 2019Hamiltonian Neural Networkscited by 712
  2. 2017DGM: A deep learning algorithm for solving partial differential equationscited by 678
  3. 2017Learning multiple visual domains with residual adapterscited by 557
  4. 2018Meta-Learning with Latent Embedding Optimizationcited by 476
  5. 2021Machine learning accelerated computational fluid dynamicscited by 471
  6. 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
  7. 2018Conditional Neural Processescited by 426
  8. 2018Meta-learning with differentiable closed-form solverscited by 397
  9. 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 379
  10. 2018Measuring the Intrinsic Dimension of Objective Landscapescited by 370
  11. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  12. 2017PDE-Net: Learning PDEs from Datacited by 311
  13. 2018How to train your MAMLcited by 309
  14. 2019Attentive Neural Processescited by 283
  15. 2018Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learningcited by 276
  16. 2019PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimizationcited by 238
  17. 2018Gradient Descent Happens in a Tiny Subspacecited by 217
  18. 2016Professor Forcing: A New Algorithm for Training Recurrent Networkscited by 216
  19. 2018Efficient parametrization of multi-domain deep neural networkscited by 168
  20. 2019Symplectic Recurrent Neural Networkscited by 150
  21. 2019Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processescited by 131
  22. 2019Meta-Learning with Warped Gradient Descentcited by 98
  23. 2020Context-aware Dynamics Model for Generalization in Model-Based Reinforcement Learningcited by 98
  24. 2017Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledgecited by 96
  25. 2020Augmenting Physical Models with Deep Networks for Complex Dynamics Forecastingcited by 95
  26. 2021Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptationcited by 50
  27. 2021LEADS: Learning Dynamical Systems that Generalize Across Environmentscited by 40
  28. 2021Meta-Learning Dynamics Forecasting Using Task Inferencecited by 38
  29. 2019Modular Meta-Learning with Shrinkagecited by 31
  30. 2020Deep learning-based reduced order models in cardiac electrophysiologycited by 22
  31. 2021Data-driven approaches for predicting spread of infectious diseases through DINNs: Disease Informed Neural Networkscited by 15

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 4; 2023: 7; 2024: 12; 2025: 13; 2026: 14.

50 papers.

Sort:

By area:

  1. 2022Continuous PDE Dynamics Forecasting with Implicit Neural Representationscited by 73
  2. 2023Generalized Teacher Forcing for Learning Chaotic Dynamicscited by 58
  3. 2023Training neural operators to preserve invariant measures of chaotic attractorscited by 46
  4. 2024Out-of-Domain Generalization in Dynamical Systems Reconstructioncited by 33
  5. 2025DISCO: learning to DISCover an evolution Operator for multi-physics-agnostic predictioncited by 21
  6. 2025True Zero-Shot Inference of Dynamical Systems Preserving Long-Term Statisticscited by 20
  7. 2024Zebra: In-Context Generative Pretraining for Solving Parametric PDEscited by 16
  8. 2024Learning Interpretable Hierarchical Dynamical Systems Models from Time Series Datacited by 15
  9. 2024Masked Autoencoders are PDE Learnerscited by 14
  10. 2024SineNet: Learning Temporal Dynamics in Time-Dependent Partial Differential Equationscited by 14
  11. 2024Strategies for Pretraining Neural Operatorscited by 14
  12. 2023Interpretable Meta-Learning of Physical Systemscited by 11
38 more
  1. 2026Test-time Generalization for Physics through Neural Operator Splittingcited by 10
  2. 2024Neural Context Flows for Meta-Learning of Dynamical Systemscited by 9
  3. 2024Meta-Dynamical State Space Models for Integrative Neural Data Analysiscited by 8
  4. 2024GEPS: Boosting Generalization in Parametric PDE Neural Solvers through Adaptive Conditioningcited by 7
  5. 2022Continual Learning of Dynamical Systems with Competitive Federated Reservoir Computingcited by 6
  6. 2023MetaPhysiCa: OOD Robustness in Physics-informed Machine Learningcited by 6
  7. 2022Phase2vec: Dynamical systems embedding with a physics-informed convolutional networkcited by 5
  8. 2025Discovering Physics Laws of Dynamical Systems via Invariant Function Learningcited by 5
  9. 2025Towards Foundational Models for Dynamical System Reconstruction: Hierarchical Meta-Learning via Mixture of Expertscited by 5
  10. 2025Weight-Space Linear Recurrent Neural Networkscited by 5
  11. 2022Towards Cross Domain Generalization of Hamiltonian Representation via Meta Learningcited by 3
  12. 2023SpReME: Sparse Regression for Multi-Environment Dynamic Systemscited by 2
  13. 2023CONFIDE: Contextual Finite Differences Modelling of PDEscited by 2
  14. 2023Learning Efficient Surrogate Dynamic Models with Graph Spline Networkscited by 2
  15. 2024Reevaluating Meta-Learning Optimization Algorithms Through Contextual Self-Modulationcited by 2
  16. 2025Tailored Forecasting from Short Time Series via Meta-learningcited by 2
  17. 2025Meta-learning Structure-Preserving Dynamicscited by 2
  18. 2025Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learningcited by 2
  19. 2025Environment Inference for Learning Generalizable Dynamical Systemcited by 2
  20. 2025Parameter-Efficient Conditioning for Material Generalization in Graph-Based Simulatorscited by 2
  21. 2026PDEInvBench: A Comprehensive Dataset and Design Space Exploration of Neural Networks for PDE Inverse Problemscited by 2
  22. 2026Sim+Real: Joint Simulation-Experiment Training Improves Balanced Prediction in Physical Systemscited by 1
  23. 2024Unsupervised Learning of Hybrid Latent Dynamics: A Learn-to-Identify Frameworkcited by 0
  24. 2024Reducing Variance in Meta-Learning via Laplace Approximation for Regression Taskscited by 0
  25. 2025Beyond Static Models: Hypernetworks for Adaptive and Generalizable Forecasting in Complex Parametric Dynamical Systemscited by 0
  26. 2025Generalizing to New Dynamical Systems via Frequency Domain Adaptationcited by 0
  27. 2025Unlocking Out-of-Distribution Generalization in Dynamics through Physics-Guided Augmentationcited by 0
  28. 2026Bayesian quantum sensing using graybox machine learningcited by 0
  29. 2026HyperKKL: Enabling Non-Autonomous State Estimation through Dynamic Weight Conditioningcited by 0
  30. 2026JEDI: Jointly Embedded Inference of Neural Dynamicscited by 0
  31. 2026Generalizing Dynamics Modeling More Easily from Representation Perspectivecited by 0
  32. 2026Bayesian Nonparametric Mixed-Effect ODEs with Gaussian Processescited by 0
  33. 2026Learning Individual Dynamics from Sparse Cross-Sectional Snapshotscited by 0
  34. 2026Learning Transferable Predictability Representationscited by 0
  35. 2026Topological Out-of-Domain Generalization in Dynamical Systems Reconstructioncited by 0
  36. 2026A Minimal Interpretable Architecture for Zero-Shot Reconstruction of Dynamical Systemscited by 0
  37. 2026Cross-Domain Pretraining for Steady-State Neural CFD Surrogatescited by 0
  38. 2026SPLICE: Sparse Physics Library Identification and Compositional Expansioncited by 0