mlsanityData to 14 Oct 2026

Multi-fidelity Reinforcement Learning Control for Complex Dynamical Systems

arXiv 2504.05588 ↗First posted 8 Apr 2025cs.LG cs.AIcited by 2

On the map: Deep reinforcement learning methods → RL theory and sample complexity

Abstract (arXiv)

Controlling instabilities in complex dynamical systems is challenging in scientific and engineering applications. Deep reinforcement learning (DRL) has seen promising results for applications in different scientific applications. The many-query nature of control tasks requires multiple interactions with real environments of the underlying physics. However, it is usually sparse to collect from the experiments or expensive to simulate for complex dynamics. Alternatively, controlling surrogate modeling could mitigate the computational cost issue. However, a fast and accurate learning-based model by offline training makes it very hard to get accurate pointwise dynamics when the dynamics are chaotic. To bridge this gap, the current work proposes a multi-fidelity reinforcement learning (MFRL) framework that leverages differentiable hybrid models for control tasks, where a physics-based hybrid model is corrected by limited high-fidelity data. We also proposed a spectrum-based reward function for RL learning. The effect of the proposed framework is demonstrated on two complex dynamics in physics. The statistics of the MFRL control result match that computed from many-query evaluations of the high-fidelity environments and outperform other SOTA baselines.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in RL theory and sample complexity, one of 15 directions in Deep reinforcement learning methods. Builds on 22 of its 50 references that are in the map. First posted on arXiv 8 Apr 2025.

Builds on

22 of 50 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024SINDy-RL: Interpretable and Efficient Model-Based Reinforcement Learning14 Mar

Other references · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  3. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,238
  4. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,209
  5. 2022MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledgecited by 478
  6. 2022Predicting Physics in Mesh-reduced Space with Temporal Attentioncited by 84
  7. 2021Scientific multi-agent reinforcement learning for wall-models of turbulent flowscited by 52
  8. 2022Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamicscited by 44
  9. 2022Learned Turbulence Modelling with Differentiable Fluid Solvers: Physics-based Loss-functions and Optimisation Horizonscited by 39
  10. 2023Debias Coarsely, Sample Conditionally: Statistical Downscaling through Optimal Transport and Probabilistic Diffusion Modelscited by 33
  11. 2019Control of chaotic systems by Deep Reinforcement Learningcited by 29
  12. 2023Bayesian Conditional Diffusion Models for Versatile Spatiotemporal Turbulence Generationcited by 29
9 more
  1. 2023Effective control of two-dimensional Rayleigh--B'enard convection: invariant multi-agent reinforcement learning is all you needcited by 27
  2. 2021Physics-informed Dyna-Style Model-Based Deep Reinforcement Learning for Dynamic Controlcited by 23
  3. 2020Multi-fidelity Generative Deep Learning Turbulent Flowscited by 21
  4. 2023Multi-fidelity reduced-order surrogate modelingcited by 18
  5. 2022Multi-fidelity reinforcement learning framework for shape optimizationcited by 9
  6. 2023Decision Stacks: Flexible Reinforcement Learning via Modular Generative Modelscited by 8
  7. 2019Restoring Chaos Using Deep Reinforcement Learningcited by 5
  8. 2022Multifidelity Reinforcement Learning with Control Variatescited by 3
  9. 2017Multi-Fidelity Reinforcement Learning with Gaussian Processescited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025Progressive multi-fidelity learning with neural networks for physical system predictionscited by 2
  2. 2026Offline Reinforcement Learning for Fluid Controls: Data-based Multi-observational Policy Extractioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Multi-Fidelity Hybrid Reinforcement Learning via Information Gain Maximization
  2. 2025Spectral Representation-based Reinforcement Learning
  3. 2025Adaptive Multi-Fidelity Reinforcement Learning for Variance Reduction in Engineering Design Optimization
  4. 2025A Multi-Fidelity Control Variate Approach for Policy Gradient Estimation
  5. 2024Interpretable and Efficient Data-driven Discovery and Control of Distributed Systems
  6. 2026Multi-Fidelity Policy Gradients Stabilize Data-Scarce Reinforcement Learning
  7. 2024Physics-informed Imitative Reinforcement Learning for Real-world Driving
  8. 2026Physics-enhanced reinforcement learning for real-time optimal control of dynamical systems