{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T15:50:16Z","timestamp":1776441016520,"version":"3.51.2"},"reference-count":171,"publisher":"Annual Reviews","issue":"1","license":[{"start":{"date-parts":[[2025,5,5]],"date-time":"2025-05-05T00:00:00Z","timestamp":1746403200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,5,5]]},"abstract":"<jats:p>Machine learning is rapidly advancing nearly every field of science and engineering, and control theory is no exception. In particular, it has shown incredible promise for handling several of the main challenges facing modern dynamics and control, including complexity, unmodeled dynamics, strong nonlinearity, and hidden variables. However, machine learning models are often expensive to train and deploy, fail to generalize beyond the training data, and suffer from a lack of explainability, interpretability, and guarantees, all of which limit their use in real-world and safety-critical control applications. Sparse nonlinear modeling and control techniques are a powerful class of machine learning that promote parsimony through sparse optimization, providing data-efficient models that are more interpretable and generalizable and have proven effective for control. In this review, we explore the use of sparse optimization in the context of machine learning to develop compact models and controllers that are easy to train, require significantly less data, and make low-latency predictions. In particular, we focus on applications in model predictive control and reinforcement learning, two of the foundational algorithms in control theory.<\/jats:p>","DOI":"10.1146\/annurev-control-030123-015238","type":"journal-article","created":{"date-parts":[[2025,1,14]],"date-time":"2025-01-14T19:40:59Z","timestamp":1736883659000},"page":"127-152","source":"Crossref","is-referenced-by-count":14,"title":["Machine Learning for Sparse Nonlinear Modeling and Control"],"prefix":"10.1146","volume":"8","author":[{"given":"Steven L.","family":"Brunton","sequence":"first","affiliation":[{"name":"1Department of Mechanical Engineering, University of Washington, Seattle, Washington, USA; email: sbrunton@uw.edu"},{"name":"2AI Institute in Dynamic Systems, University of Washington, Seattle, Washington, USA"},{"name":"3Department of Applied Mathematics, University of Washington, Seattle, Washington, USA"}]},{"given":"Nicholas","family":"Zolman","sequence":"additional","affiliation":[{"name":"1Department of Mechanical Engineering, University of Washington, Seattle, Washington, USA; email: sbrunton@uw.edu"},{"name":"2AI Institute in Dynamic Systems, University of Washington, Seattle, Washington, USA"}]},{"given":"J. Nathan","family":"Kutz","sequence":"additional","affiliation":[{"name":"2AI Institute in Dynamic Systems, University of Washington, Seattle, Washington, USA"},{"name":"3Department of Applied Mathematics, University of Washington, Seattle, Washington, USA"},{"name":"4Department of Electrical and Computer Engineering, University of Washington, Seattle, Washington, USA"},{"name":"1Department of Mechanical Engineering, University of Washington, Seattle, Washington, USA; email: sbrunton@uw.edu"}]},{"given":"Urban","family":"Fasel","sequence":"additional","affiliation":[{"name":"1Department of Mechanical Engineering, University of Washington, Seattle, Washington, USA; email: sbrunton@uw.edu"},{"name":"2AI Institute in Dynamic Systems, University of Washington, Seattle, Washington, USA"},{"name":"5Department of Aeronautics, Imperial College London, London, United Kingdom"}]}],"member":"22","reference":[{"issue":"3","key":"B1","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1016\/0005-1098(89)90002-2","article-title":"Model predictive control: theory and practice\u2014a survey","volume":"25","year":"1989","journal-title":"Automatica"},{"issue":"4","key":"B2","first-page":"667","article-title":"Model predictive control: past, present and future","volume":"23","year":"1999","journal-title":"Comput. Chem. Eng."},{"issue":"12","key":"B3","doi-asserted-by":"crossref","first-page":"2967","DOI":"10.1016\/j.automatica.2014.10.128","article-title":"Model predictive control: recent developments and future promise","volume":"50","year":"2014","journal-title":"Automatica"},{"key":"B4","volume-title":"Reinforcement Learning: An Introduction","year":"1998"},{"key":"B5","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1146\/annurev-control-053018-023825","article-title":"A tour of reinforcement learning: the view from continuous control","volume":"2","year":"2019","journal-title":"Annu. Rev. Control Robot. Auton. Syst."},{"key":"B6","doi-asserted-by":"crossref","first-page":"477","DOI":"10.1146\/annurev-fluid-010719-060214","article-title":"Machine learning for fluid mechanics","volume":"52","year":"2020","journal-title":"Annu. Rev. Fluid Mech."},{"key":"B7","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1017\/jfm.2017.823","article-title":"Constrained sparse Galerkin regression","volume":"838","year":"2018","journal-title":"J. Fluid Mech."},{"key":"B8","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1146\/annurev-fluid-010518-040547","article-title":"Turbulence modeling in the age of data","volume":"51","year":"2019","journal-title":"Annu. Rev. Fluid Mech."},{"key":"B9","doi-asserted-by":"crossref","first-page":"339","DOI":"10.1007\/s00162-020-00536-w","article-title":"Data-driven modeling of the chaotic thermal convection in an annular thermosyphon","volume":"34","year":"2020","journal-title":"Theor. Comput. Fluid Dyn."},{"key":"B10","doi-asserted-by":"crossref","first-page":"A37","DOI":"10.1017\/jfm.2019.959","article-title":"Low-order model for successive bifurcations of the fluidic pinball","volume":"884","year":"2020","journal-title":"J. Fluid Mech."},{"issue":"2","key":"B11","first-page":"579","article-title":"Discovery of algebraic Reynolds-stress models using sparse symbolic regression","volume":"104","year":"2020","journal-title":"Flow Turbul. Combust."},{"issue":"17","key":"B12","doi-asserted-by":"crossref","first-page":"e2020GL088376","DOI":"10.1029\/2020GL088376","article-title":"Data-driven equation discovery of ocean mesoscale closures","volume":"47","year":"2020","journal-title":"Geophys. Res. Lett."},{"key":"B13","doi-asserted-by":"crossref","first-page":"A11","DOI":"10.1017\/jfm.2021.53","article-title":"Sparse identification of multiphase turbulence closures for coupled fluid\u2013particle flows","volume":"914","year":"2021","journal-title":"J. Fluid Mech."},{"issue":"5","key":"B14","doi-asserted-by":"crossref","first-page":"1936","DOI":"10.1523\/JNEUROSCI.16-05-01936.1996","article-title":"A framework for mesencephalic dopamine systems based on predictive Hebbian learning","volume":"16","year":"1996","journal-title":"J. Neurosci."},{"key":"B15","volume-title":"Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems","year":"2001"},{"key":"B16","first-page":"449","article-title":"Effective construction of linear state-variable models from input\/output data","year":"1965","journal-title":"Proceedings of the 3rd Allerton Conference on Circuit and System Theory"},{"issue":"5","key":"B17","doi-asserted-by":"crossref","first-page":"620","DOI":"10.2514\/3.20031","article-title":"An eigensystem realization algorithm for modal parameter identification and model reduction","volume":"8","year":"1985","journal-title":"J. Guid. Control Dyn."},{"key":"B18","volume-title":"Applied System Identification","year":"1994"},{"key":"B19","volume-title":"System Identification: Theory for the User","year":"1999"},{"key":"B20","volume-title":"Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control","year":"2022"},{"issue":"6245","key":"B21","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1126\/science.aaa8415","article-title":"Machine learning: trends, perspectives, and prospects","volume":"349","year":"2015","journal-title":"Science"},{"key":"B22","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1016\/j.arcontrol.2021.10.001","article-title":"Contraction theory for nonlinear stability analysis and learning-based control: a tutorial overview","volume":"52","year":"2021","journal-title":"Annu. Rev. Control"},{"key":"B23","first-page":"571","article-title":"Distributed neural network control with dependability guarantees: a compositional port-Hamiltonian approach","year":"2022","journal-title":"Proceedings of the 4th Annual Learning for Dynamics and Control Conference"},{"key":"B24","first-page":"91","article-title":"Learning over all stabilizing nonlinear controllers for a partially-observed linear system","volume":"7","year":"2022","journal-title":"IEEE Control Syst. Lett."},{"key":"B25","first-page":"2765","article-title":"Neural system level synthesis: learning over all stabilizing policies for nonlinear systems","year":"2022","journal-title":"2022 IEEE 61st Conference on Decision and Control"},{"key":"B26","first-page":"5385","article-title":"Recurrent neural network controllers synthesis with stability guarantees for partially observed systems","volume":"36","year":"2022","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"B27","article-title":"Learning stable and passive neural differential equations","year":"2024"},{"key":"B28","article-title":"On robust reinforcement learning with Lipschitz-bounded policy networks","year":"2024"},{"key":"B29","article-title":"Optimal distributed control with stability guarantees by training a network of neural closed-loop maps","year":"2024"},{"key":"B30","volume-title":"Feedback Control Theory","year":"2013"},{"issue":"15","key":"B31","doi-asserted-by":"crossref","first-page":"3932","DOI":"10.1073\/pnas.1517384113","article-title":"Discovering governing equations from data by sparse identification of nonlinear dynamical systems","volume":"113","year":"2016","journal-title":"PNAS"},{"key":"B32","doi-asserted-by":"crossref","first-page":"e1602614","DOI":"10.1126\/sciadv.1602614","article-title":"Data-driven discovery of partial differential equations","volume":"3","year":"2017","journal-title":"Sci. Adv."},{"issue":"24","key":"B33","doi-asserted-by":"crossref","first-page":"9943","DOI":"10.1073\/pnas.0609476104","article-title":"Automated reverse engineering of nonlinear dynamical systems","volume":"104","year":"2007","journal-title":"PNAS"},{"issue":"5923","key":"B34","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1126\/science.1165893","volume":"324","year":"2009","journal-title":"Science"},{"key":"B35","article-title":"Interpretable machine learning for science with PySR and SymbolicRegression.jl","year":"2023"},{"key":"B36","doi-asserted-by":"crossref","first-page":"116072","DOI":"10.1016\/j.cma.2023.116072","article-title":"Reduced order modeling of parametrized systems through autoencoders and SINDy approach: continuation of periodic solutions","volume":"411","year":"2023","journal-title":"Comput. Methods Appl. Mech. Eng."},{"key":"B37","doi-asserted-by":"crossref","first-page":"094401","DOI":"10.1103\/PhysRevFluids.6.094401","article-title":"Promoting global stability in data-driven models of quadratic nonlinear dynamics","volume":"6","year":"2021","journal-title":"Phys. Rev. Fluids"},{"key":"B38","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1017\/S0022112010001217","article-title":"Dynamic mode decomposition of numerical and experimental data","volume":"656","year":"2010","journal-title":"J. Fluid Mech."},{"key":"B39","first-page":"115","article-title":"Spectral analysis of nonlinear flows","volume":"645","year":"2009","journal-title":"J. Fluid Mech."},{"key":"B40","volume-title":"Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems","year":"2016"},{"key":"B41","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1146\/annurev-fluid-030121-015835","article-title":"Dynamic mode decomposition and its variants","volume":"54","year":"2022","journal-title":"Annu. Rev. Fluid Mech."},{"issue":"2","key":"B42","doi-asserted-by":"crossref","first-page":"229","DOI":"10.1137\/21M1401243","article-title":"Modern Koopman theory for dynamical systems","volume":"64","year":"2022","journal-title":"SIAM Rev"},{"issue":"2219","key":"B43","doi-asserted-by":"crossref","first-page":"20180335","DOI":"10.1098\/rspa.2018.0335","article-title":"Sparse identification of nonlinear dynamics for model predictive control in the low-data limit","volume":"474","year":"2018","journal-title":"Proc. R. Soc. A"},{"key":"B44","article-title":"SINDy-RL: interpretable and efficient model-based reinforcement learning","year":"2024"},{"key":"B45","article-title":"Automatic differentiation in PyTorch","volume-title":"NIPS 2017 Autodiff Workshop: The Future of Gradient-Based Machine Learning Software and Techniques","year":"2017"},{"issue":"153","key":"B46","first-page":"1","article-title":"Automatic differentiation in machine learning: a survey","volume":"18","year":"2018","journal-title":"J. Mach. Learn. Res."},{"key":"B47","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1016\/j.jprocont.2022.06.001","article-title":"Differentiable predictive control: deep learning alternative to explicit model predictive control for unknown nonlinear systems","volume":"116","year":"2022","journal-title":"J. Process Control"},{"key":"B48","first-page":"8289","article-title":"Differentiable MPC for end-to-end planning and control","volume-title":"Advances in Neural Information Processing Systems 31","year":"2018"},{"key":"B49","first-page":"7178","article-title":"End-to-end differentiable physics for learning and control","volume-title":"Advances in Neural Information Processing Systems 31","year":"2018"},{"key":"B50","article-title":"Learning to control PDEs with differentiable physics","year":"2020"},{"key":"B51","first-page":"7979","article-title":"Pontryagin differentiable programming: an end-to-end learning and control framework","volume-title":"Advances in Neural Information Processing Systems 33","year":"2020"},{"issue":"1\u20134","key":"B52","first-page":"213","article-title":"Optimum aerodynamic design using the Navier\u2013Stokes equations","volume":"10","year":"1998","journal-title":"Theor. Comput. Fluid Dyn."},{"key":"B53","volume-title":"Nonlinear System Identification: NARMAX Methods in the Time, Frequency, and Spatio-Temporal Domains","year":"2013"},{"issue":"1","key":"B54","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regression shrinkage and selection via the lasso","volume":"58","year":"1996","journal-title":"J. R. Stat. Soc. B"},{"issue":"3","key":"B55","doi-asserted-by":"crossref","first-page":"948","DOI":"10.1137\/18M1189828","article-title":"On the convergence of the SINDy algorithm","volume":"17","year":"2019","journal-title":"Multiscale Model. Simul."},{"key":"B56","article-title":"PySINDy: a comprehensive Python package for robust sparse system identification","year":"2021"},{"issue":"1","key":"B57","first-page":"63","article-title":"Application de la th\u00e9orie des \u00e9quations int\u00e9grales lin\u00e9aires aux syst\u00e9mes d'\u00e9quations diff\u00e9rentielles non lin\u00e9aires","volume":"59","year":"1932","journal-title":"Acta Math"},{"key":"B58","doi-asserted-by":"crossref","first-page":"196865","DOI":"10.1109\/ACCESS.2020.3034077","article-title":"Numerical differentiation of noisy data: a unifying multi-objective optimization framework","volume":"8","year":"2020","journal-title":"IEEE Access"},{"key":"B59","article-title":"EKF-SINDy: empowering the extended Kalman filter with sparse identification of nonlinear dynamics","year":"2024"},{"issue":"1","key":"B60","doi-asserted-by":"crossref","first-page":"010203","DOI":"10.1103\/PhysRevE.101.010203","article-title":"Using noisy or incomplete data to discover models of spatiotemporal dynamics","volume":"101","year":"2020","journal-title":"Phys. Rev. E"},{"issue":"3","key":"B61","doi-asserted-by":"crossref","first-page":"1474","DOI":"10.1137\/20M1343166","volume":"19","year":"2021","journal-title":"Multiscale Model. Simul."},{"issue":"2260","key":"B62","doi-asserted-by":"crossref","first-page":"20210904","DOI":"10.1098\/rspa.2021.0904","article-title":"Ensemble-SINDy: robust sparse model discovery in the low-data, high-noise limit, with active learning and control","volume":"478","year":"2022","journal-title":"Proc. R. Soc. A"},{"issue":"2","key":"B63","doi-asserted-by":"crossref","first-page":"023302","DOI":"10.1103\/PhysRevE.96.023302","article-title":"Sparse model selection via integral terms","volume":"96","year":"2017","journal-title":"Phys. Rev. E"},{"key":"B64","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1023\/A:1018054314350","article-title":"Bagging predictors","volume":"24","year":"1996","journal-title":"Mach. Learn."},{"key":"B65","doi-asserted-by":"crossref","first-page":"985","DOI":"10.1007\/978-3-642-21551-3_33","article-title":"Bagging, boosting and ensemble methods","volume-title":"Handbook of Computational Statistics: Concepts and Methods","year":"2012"},{"key":"B66","volume-title":"The Elements of Statistical Learning: Data Mining, Inference, and Prediction","year":"2009"},{"issue":"4","key":"B67","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1111\/j.1467-9868.2010.00740.x","article-title":"Stability selection","volume":"72","year":"2010","journal-title":"J. R. Stat. Soc. B"},{"key":"B68","article-title":"Convergence of uncertainty estimates in ensemble and Bayesian sparse model discovery","year":"2023"},{"issue":"1","key":"B69","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1137\/15M1013857","article-title":"Dynamic mode decomposition with control","volume":"15","year":"2016","journal-title":"SIAM J. Appl. Dyn. Syst."},{"issue":"1","key":"B70","doi-asserted-by":"crossref","first-page":"380","DOI":"10.1137\/M1124176","article-title":"Variable projection methods for an optimized dynamic mode decomposition","volume":"17","year":"2018","journal-title":"SIAM J. Appl. Dyn. Syst."},{"issue":"2229","key":"B71","doi-asserted-by":"crossref","first-page":"20210199","DOI":"10.1098\/rsta.2021.0199","article-title":"Bagging, optimized dynamic mode decomposition for robust, stable forecasting with spatial and temporal uncertainty quantification","volume":"380","year":"2022","journal-title":"Philos. Trans. R. Soc. A"},{"key":"B72","first-page":"6492","article-title":"Koopman operator based observer synthesis for control-affine nonlinear systems","year":"2016","journal-title":"2016 55th IEEE Conference on Decision and Control"},{"key":"B73","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1016\/j.automatica.2019.05.016","article-title":"Koopman operator-based model reduction for switched-system control of PDEs","volume":"106","year":"2019","journal-title":"Automatica"},{"issue":"3","key":"B74","doi-asserted-by":"crossref","first-page":"2162","DOI":"10.1137\/20M1325678","article-title":"Data-driven model predictive control using interpolated Koopman generators","volume":"19","year":"2020","journal-title":"SIAM J. Appl. Dyn. Syst."},{"key":"B75","doi-asserted-by":"crossref","first-page":"132416","DOI":"10.1016\/j.physd.2020.132416","article-title":"Data-driven approximation of the Koopman generator: model reduction, system identification, and control","volume":"406","year":"2020","journal-title":"Phys. D"},{"issue":"6","key":"B76","first-page":"2012","article-title":"Data-driven safety-critical control: synthesizing control barrier functions with Koopman operators","volume":"5","year":"2020","journal-title":"IEEE Control Syst. Lett."},{"key":"B77","first-page":"7350","article-title":"Koopman NMPC: Koopman-based learning and nonlinear model predictive control of control-affine systems","year":"2021","journal-title":"2021 IEEE International Conference on Robotics and Automation"},{"key":"B78","article-title":"Quadrotor trajectory tracking with learned dynamics: joint Koopman-based learning of system models and function dictionaries","year":"2021"},{"key":"B79","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1016\/j.arcontrol.2021.09.002","article-title":"Koopman operator dynamical models: learning, analysis and control","volume":"52","year":"2021","journal-title":"Annu. Rev. Control"},{"issue":"15","key":"B80","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1016\/j.ifacol.2022.07.601","article-title":"Towards data-driven LQR with Koopmanizing flows","volume":"55","year":"2022","journal-title":"IFAC-PapersOnLine"},{"key":"B81","article-title":"Koopman constrained policy optimization: a Koopman operator theoretic method for differentiable optimal control in robotics","volume-title":"ICML 2023 Workshop on Differentiable Almost Everything: Differentiable Relaxations, Algorithms, Operators, and Simulators","year":"2023"},{"key":"B82","article-title":"Koopman-assisted reinforcement learning","volume-title":"NeurIPS 2023 AI for Science Workshop","year":"2023"},{"key":"B83","first-page":"366","article-title":"Detecting strange attractors in turbulence","year":"1981","journal-title":"Dynamical Systems and Turbulence"},{"issue":"3","key":"B84","doi-asserted-by":"crossref","first-page":"812","DOI":"10.1016\/j.camwa.2010.05.028","article-title":"Relationship between singular spectrum analysis and Fourier analysis: theory and application to the monitoring of volcanic activity","volume":"60","year":"2010","journal-title":"Comput. Math. Appl."},{"key":"B85","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1038\/s41467-017-00030-8","article-title":"Chaos as an intermittently forced linear system","volume":"8","year":"2017","journal-title":"Nat. Commun."},{"key":"B86","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.camwa.2023.08.016","article-title":"Uncertainty quantification for nonlinear solid mechanics using reduced order models with Gaussian process regression","volume":"149","year":"2023","journal-title":"Comput. Math. Appl."},{"key":"B87","first-page":"465","article-title":"PILCO: a model-based and data-efficient approach to policy search","year":"2011","journal-title":"Proceedings of the 28th International Conference on Machine Learning"},{"issue":"5","key":"B88","first-page":"1327","article-title":"Review on model predictive control: an engineering perspective","volume":"117","year":"2021","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"B89","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1146\/annurev-control-090419-075625","article-title":"Learning-based model predictive control: toward safe learning in control","volume":"3","year":"2020","journal-title":"Annu. Rev. Control Robot. Auton. Syst."},{"key":"B90","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1146\/annurev-control-042920-020211","article-title":"Safe learning in robotics: from learning-based control to safe reinforcement learning","volume":"5","year":"2022","journal-title":"Annu. Rev. Control Robot. Auton. Syst."},{"key":"B91","article-title":"DeepMPC: learning deep latent features for model predictive control","volume-title":"Robotics: Science and Systems XI","year":"2015"},{"key":"B92","first-page":"528","article-title":"Learning deep control policies for autonomous aerial vehicles with MPC-guided policy search","year":"2016","journal-title":"2016 IEEE International Conference on Robotics and Automation"},{"key":"B93","article-title":"Aggressive deep driving: model predictive control with a cnn cost model","year":"2017"},{"key":"B94","article-title":"Deep learning tubes for tube MPC","year":"2020"},{"issue":"1","key":"B95","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1007\/BF02532251","article-title":"Fitting autoregressive models for prediction","volume":"21","year":"1969","journal-title":"Ann. Inst. Stat. Math."},{"issue":"45","key":"B96","doi-asserted-by":"crossref","first-page":"22445","DOI":"10.1073\/pnas.1906995116","article-title":"Data-driven discovery of coordinates and governing equations","volume":"116","year":"2019","journal-title":"PNAS"},{"issue":"2276","key":"B97","doi-asserted-by":"crossref","first-page":"20230422","DOI":"10.1098\/rspa.2023.0422","article-title":"Discovering governing equations from partial measurements with deep delay autoencoders","volume":"479","year":"2023","journal-title":"Proc. R. Soc. A"},{"issue":"6","key":"B98","doi-asserted-by":"crossref","first-page":"063116","DOI":"10.1063\/1.5027470","article-title":"Sparse identification of nonlinear dynamics for rapid model recovery","volume":"28","year":"2018","journal-title":"Chaos"},{"key":"B99","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1016\/j.arcontrol.2017.11.001","article-title":"Stochastic model predictive control with active uncertainty learning: a survey on dual control","volume":"45","year":"2018","journal-title":"Annu. Rev. Control"},{"issue":"2217","key":"B100","doi-asserted-by":"crossref","first-page":"20180305","DOI":"10.1098\/rspa.2018.0305","article-title":"Robust data-driven discovery of governing physical laws with error bars","volume":"474","year":"2018","journal-title":"Proc. R. Soc. A"},{"key":"B101","article-title":"Dynamical system identification, model selection and model uncertainty quantification by Bayesian inference","year":"2024"},{"key":"B102","article-title":"Rapid Bayesian identification of sparse nonlinear dynamics from scarce and noisy data","year":"2024"},{"key":"B103","volume-title":"The Behavior of Organisms: An Experimental Analysis","year":"2019 (1938)"},{"key":"B104","volume-title":"Experimental Psychology","year":"2008 (1938)"},{"issue":"7540","key":"B105","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1038\/nature14236","article-title":"Human-level control through deep reinforcement learning","volume":"518","year":"2015","journal-title":"Nature"},{"issue":"7676","key":"B106","doi-asserted-by":"crossref","first-page":"354","DOI":"10.1038\/nature24270","article-title":"Mastering the game of Go without human knowledge","volume":"550","year":"2017","journal-title":"Nature"},{"issue":"7782","key":"B107","doi-asserted-by":"crossref","first-page":"350","DOI":"10.1038\/s41586-019-1724-z","article-title":"Grandmaster level in Starcraft II using multi-agent reinforcement learning","volume":"575","year":"2019","journal-title":"Nature"},{"issue":"7976","key":"B108","doi-asserted-by":"crossref","first-page":"982","DOI":"10.1038\/s41586-023-06419-4","article-title":"Champion-level drone racing using deep reinforcement learning","volume":"620","year":"2023","journal-title":"Nature"},{"issue":"7897","key":"B109","doi-asserted-by":"crossref","first-page":"414","DOI":"10.1038\/s41586-021-04301-9","article-title":"Magnetic control of tokamak plasmas through deep reinforcement learning","volume":"602","year":"2022","journal-title":"Nature"},{"issue":"7","key":"B110","doi-asserted-by":"crossref","first-page":"eaap7885","DOI":"10.1126\/sciadv.aap7885","article-title":"Deep reinforcement learning for de novo drug design","volume":"4","year":"2018","journal-title":"Sci. Adv."},{"issue":"15","key":"B111","doi-asserted-by":"crossref","first-page":"158004","DOI":"10.1103\/PhysRevLett.118.158004","article-title":"Flow navigation by smart microswimmers via reinforcement learning","volume":"118","year":"2017","journal-title":"Phys. Rev. Lett."},{"key":"B112","first-page":"1404","article-title":"A continuous reinforcement learning strategy for closed-loop control in fluid dynamics","year":"2017","journal-title":"35th AIAA Applied Aerodynamics Conference"},{"issue":"23","key":"B113","doi-asserted-by":"crossref","first-page":"5849","DOI":"10.1073\/pnas.1800923115","article-title":"Efficient collective swimming by harnessing vortices through deep reinforcement learning","volume":"115","year":"2018","journal-title":"PNAS"},{"issue":"42","key":"B114","doi-asserted-by":"crossref","first-page":"26091","DOI":"10.1073\/pnas.2004939117","article-title":"Reinforcement learning for bluff body active flow control in experiments and simulations","volume":"117","year":"2020","journal-title":"PNAS"},{"key":"B115","first-page":"368","article-title":"Deep reinforcement learning applied to active flow control","volume-title":"Data-Driven Fluid Mechanics: Combining First Principles and Machine Learning","year":"2023"},{"issue":"1","key":"B116","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1038\/s42256-020-00272-0","article-title":"Automating turbulence modelling by multi-agent reinforcement learning","volume":"3","year":"2021","journal-title":"Nat. Mach. Intell."},{"key":"B117","doi-asserted-by":"crossref","first-page":"1443","DOI":"10.1038\/s41467-022-28957-7","article-title":"Scientific multi-agent reinforcement learning for wall-models of turbulent flows","volume":"13","year":"2022","journal-title":"Nat. Commun."},{"issue":"2","key":"B118","doi-asserted-by":"crossref","first-page":"62","DOI":"10.3390\/fluids7020062","article-title":"Flow control in wings and discovery of novel approaches via deep reinforcement learning","volume":"7","year":"2022","journal-title":"Fluids"},{"issue":"3","key":"B119","doi-asserted-by":"crossref","first-page":"031301","DOI":"10.1063\/5.0143913","article-title":"Recent advances in applying deep reinforcement learning for flow control: perspectives and future directions","volume":"35","year":"2023","journal-title":"Phys. Fluids"},{"key":"B120","article-title":"Offline reinforcement learning: tutorial, review, and perspectives on open problems","year":"2020"},{"issue":"2","key":"B121","first-page":"21","article-title":"Imitation learning: a survey of learning methods","volume":"50","year":"2017","journal-title":"ACM Comput. Surv."},{"key":"B122","doi-asserted-by":"crossref","first-page":"362","DOI":"10.1007\/s41315-019-00103-5","article-title":"Survey of imitation learning for robotic manipulation","volume":"3","year":"2019","journal-title":"Int. J. Intell. Robot. Appl."},{"issue":"9","key":"B123","doi-asserted-by":"crossref","first-page":"14128","DOI":"10.1109\/TITS.2022.3144867","article-title":"A survey on imitation learning techniques for end-to-end autonomous vehicles","volume":"23","year":"2022","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"B124","year":"2016","journal-title":"Prioritized experience replay"},{"key":"B125","first-page":"350","article-title":"Experience replay for continual learning","volume-title":"Advances in Neural Information Processing Systems 32","year":"2019"},{"key":"B126","first-page":"5049","article-title":"Hindsight experience replay","volume-title":"Advances in Neural Information Processing Systems 30","year":"2017"},{"issue":"11","key":"B127","doi-asserted-by":"crossref","first-page":"13344","DOI":"10.1109\/TPAMI.2023.3292075","article-title":"Transfer learning in deep reinforcement learning: a survey","volume":"45","year":"2023","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B128","article-title":"Learning to reinforcement learn","year":"2016"},{"key":"B129","first-page":"1126","article-title":"Model-agnostic meta-learning for fast adaptation of deep networks","year":"2017","journal-title":"Proceedings of the 34th International Conference on Machine Learning"},{"key":"B130","article-title":"Benchmarking model-based reinforcement learning","year":"2019"},{"issue":"1","key":"B131","first-page":"497","article-title":"A survey of numerical methods for optimal control","volume":"135","year":"2009","journal-title":"Adv. Astronaut. Sci."},{"key":"B132","first-page":"4906","article-title":"Synthesis and stabilization of complex behaviors through online trajectory optimization","year":"2012","journal-title":"2012 IEEE\/RSJ International Conference on Intelligent Robots and Systems"},{"key":"B133","first-page":"2944","article-title":"Learning continuous control policies by stochastic value gradients","volume-title":"Advances in Neural Information Processing Systems 28","year":"2015"},{"issue":"1","key":"B134","first-page":"1334","article-title":"End-to-end training of deep visuomotor policies","volume":"17","year":"2016","journal-title":"J. Mach. Learn. Res."},{"key":"B135","first-page":"1332","article-title":"Sample efficient reinforcement learning with Gaussian processes","year":"2014","journal-title":"Proceedings of the 31st International Conference on Machine Learning"},{"key":"B136","first-page":"528","article-title":"Dyna-style planning with linear function approximation and prioritized sweeping","year":"2008","journal-title":"In UAI'08: Proceedings of the Twenty-Fourth Conference on Uncertainty in Artificial Intelligence"},{"key":"B137","first-page":"4754","article-title":"Deep reinforcement learning in a handful of trials using probabilistic dynamics models","volume-title":"Advances in Neural Information Processing Systems 31","year":"2018"},{"key":"B138","first-page":"7559","article-title":"Neural network dynamics for model-based deep reinforcement learning with model-free fine-tuning","year":"2018","journal-title":"2018 IEEE International Conference on Robotics and Automation"},{"key":"B139","article-title":"Dream to control: learning behaviors by latent imagination","year":"2020","journal-title":"2020 International Conference on Learning Representations"},{"key":"B140","article-title":"Model-based reinforcement learning with SINDy","year":"2022"},{"key":"B141","first-page":"216","article-title":"Integrated architectures for learning, planning, and reacting based on approximating dynamic programming","year":"1990","journal-title":"Machine Learning: Proceedings of the Seventh International Conference on Machine Learning"},{"key":"B142","first-page":"617","article-title":"Model-based reinforcement learning via meta-policy optimization","year":"2018","journal-title":"Proceedings of the 2nd Conference on Robot Learning"},{"issue":"9","key":"B143","first-page":"4166","article-title":"Integration of neural network-based symbolic regression in deep learning for scientific discovery","volume":"32","year":"2020","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B144","first-page":"4442","article-title":"Learning equations for extrapolation and control","year":"2018","journal-title":"Proceedings of the 35th International Conference on Machine Learning"},{"key":"B145","first-page":"1800","article-title":"Simple random search of static linear policies is competitive for reinforcement learning","volume-title":"Advances in Neural Information Processing Systems 31","year":"2018"},{"key":"B146","first-page":"6551","article-title":"Towards generalization and simplicity in continuous control","volume-title":"Advances in Neural Information Processing Systems 30","year":"2017"},{"key":"B147","doi-asserted-by":"crossref","first-page":"968305","DOI":"10.3389\/frobt.2022.968305","article-title":"NN-Poly: approximating common neural networks with Taylor polynomials to imbue dynamical system constraints","volume":"9","year":"2022","journal-title":"Front. Robot. AI"},{"key":"B148","article-title":"Parametric PDE control with deep reinforcement learning and differentiable L0-sparse polynomial policies","year":"2024"},{"key":"B149","article-title":"Online identification and control of PDEs via reinforcement learning methods","year":"2023"},{"issue":"2","key":"B150","doi-asserted-by":"crossref","first-page":"8142","DOI":"10.1016\/j.ifacol.2020.12.2299","article-title":"Structured online learning-based control of continuous-time nonlinear systems","volume":"53","year":"2020","journal-title":"IFAC-PapersOnLine"},{"key":"B151","doi-asserted-by":"crossref","first-page":"4950","DOI":"10.1038\/s41467-018-07210-0","article-title":"Deep learning for universal linear embeddings of nonlinear dynamics","volume":"9","year":"2018","journal-title":"Nat. Commun."},{"issue":"7","key":"B152","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1038\/s43588-022-00281-6","article-title":"Automated discovery of fundamental variables hidden in experimental data","volume":"2","year":"2022","journal-title":"Nat. Comput. Sci."},{"issue":"1","key":"B153","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1109\/TAC.1981.1102568","article-title":"Principal component analysis in linear systems: controllability, observability, and model reduction","volume":"26","year":"1981","journal-title":"IEEE Trans. Autom. Control"},{"key":"B154","volume-title":"Nonlinear Oscillations, Dynamical Systems, and Bifurcations of Vector Fields","year":"1983"},{"key":"B155","article-title":"Learning deep neural network representations for Koopman operators of nonlinear dynamical systems","year":"2017"},{"key":"B156","first-page":"1130","article-title":"Learning Koopman invariant subspaces for dynamic mode decomposition","volume-title":"Advances in Neural Information Processing Systems 30","year":"2017"},{"key":"B157","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1038\/s41467-017-02388-1","article-title":"VAMPnets: deep learning of molecular kinetics","volume":"9","year":"2018","journal-title":"Nat. Commun."},{"issue":"1","key":"B158","doi-asserted-by":"crossref","first-page":"558","DOI":"10.1137\/18M1177846","article-title":"Linearly-recurrent autoencoder networks for learning dynamics","volume":"18","year":"2019","journal-title":"SIAM J. Appl. Dyn. Syst."},{"key":"B159","doi-asserted-by":"crossref","first-page":"132401","DOI":"10.1016\/j.physd.2020.132401","article-title":"Lift & learn: physics-informed machine learning for large-scale nonlinear dynamical systems","volume":"406","year":"2020","journal-title":"Phys. D"},{"issue":"1","key":"B160","doi-asserted-by":"crossref","first-page":"982","DOI":"10.1137\/23M1561129","article-title":"Exact and optimal quadratization of nonlinear finite-dimensional nonautonomous dynamical systems","volume":"23","year":"2024","journal-title":"SIAM J. Appl. Dyn. Syst."},{"key":"B161","doi-asserted-by":"crossref","first-page":"521","DOI":"10.1146\/annurev-fluid-121021-025220","article-title":"Learning nonlinear reduced models from data with operator inference","volume":"56","year":"2024","journal-title":"Annu. Rev. Fluid Mech."},{"issue":"2","key":"B162","doi-asserted-by":"crossref","first-page":"981","DOI":"10.1137\/20M1364849","article-title":"Stability domains for quadratic-bilinear reduced-order models","volume":"20","year":"2021","journal-title":"SIAM J. Appl. Dyn. Syst."},{"key":"B163","article-title":"Gauge equivariant mesh CNNs: anisotropic convolutions on geometric graphs","year":"2020"},{"key":"B164","article-title":"Incorporating symmetry into deep dynamics models for improved generalization","year":"2020"},{"key":"B165","article-title":"Relevance of rotationally equivariant convolutions for predicting molecular properties","year":"2020"},{"key":"B166","doi-asserted-by":"crossref","first-page":"2453","DOI":"10.1038\/s41467-022-29939-5","article-title":"E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials","volume":"13","year":"2022","journal-title":"Nat. Commun."},{"key":"B167","first-page":"2241","article-title":"Lie point symmetry data augmentation for neural PDE solvers","year":"2022","journal-title":"Proceedings of the 39th International Conference on Machine Learning"},{"key":"B168","article-title":"A unified framework to enforce, discover, and promote symmetry in machine learning","year":"2023"},{"key":"B169","article-title":"Symmetry-informed governing equation discovery","year":"2024"},{"key":"B170","article-title":"The bitter lesson","volume-title":"Incomplete Ideas","year":"2019"},{"key":"B171","first-page":"16","article-title":"SINDy with control: a tutorial","year":"2021","journal-title":"2021 60th IEEE Conference on Decision and Control"}],"container-title":["Annual Review of Control, Robotics, and Autonomous Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.annualreviews.org\/content\/journals\/10.1146\/annurev-control-030123-015238?crawler=true&mimetype=application\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,14]],"date-time":"2026-03-14T04:15:04Z","timestamp":1773461704000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.annualreviews.org\/content\/journals\/10.1146\/annurev-control-030123-015238"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,5]]},"references-count":171,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2025,5,5]]}},"URL":"https:\/\/doi.org\/10.1146\/annurev-control-030123-015238","relation":{},"ISSN":["2573-5144"],"issn-type":[{"value":"2573-5144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,5]]}}}