Abstract
This work offers a discussion on how computational mechanics and physics-informed machine learning can be integrated into the process of sensing, understanding, and reasoning of physical phenomena. A foundation in physics can leverage interpretability, data efficiency, and generalization of the models sought for the dynamics of complex physical systems. Consequently, this synergy results in promising approaches to develop world models that are capable of performing accurate and reliable simulations (reasoning) in low-data regimes. Among the possible alternative formulations, we highlight how thermodynamics offers a general framework to construct inductive biases, demonstrating its potential in applications where physics-consistent predictions are essential.
Similar content being viewed by others
Data Availability
Not applicable.
Code Availability
Not applicable.
References
Tamkin A, Brundage M, Clark J, Ganguli D (2021) Understanding the capabilities, limitations, and societal impact of large language models. arXiv preprint arXiv:2102.02503
Li K, Hopkins AK, Bau D, Viégas F, Pfister H, Wattenberg M (2022) Emergent world representations: exploring a sequence model trained on a synthetic task. arXiv preprint arXiv:2210.13382
Lake BM, Ullman TD, Tenenbaum JB, Gershman SJ (2017) Building machines that learn and think like people. Behav Brain Sci 40:253
Kubricht JR, Holyoak KJ, Lu H (2017) Intuitive physics: current research and controversies. Trends Cogn Sci 21(10):749–759
McCloskey M (1983) Intuitive physics. Sci Am 248(4):122–131
Reynolds CR, Fletcher-Janzen E (2007) Encyclopedia of special education: a reference for the education of children, adolescents, and adults with disabilities and other exceptional individuals, vol 3. Wiley, New York
Piloto LS, Weinstein A, Battaglia P, Botvinick M (2022) Intuitive physics learning in a deep-learning model inspired by developmental psychology. Nature human behaviour 6(9):1257–1267
Allen KR, Lopez-Guevara T, Stachenfeld K, Sanchez-Gonzalez A, Battaglia P, Hamrick J, Pfaff T (2022) Physical design using differentiable learned simulators. arXiv preprint arXiv:2202.00728
Sanchez-Gonzalez A, Godwin J, Pfaff T, Ying R, Leskovec J, Battaglia P (2020) Learning to simulate complex physics with graph networks. In: International conference on machine learning, pp 8459–8468. PMLR
Hernandez Q, Badias A, Gonzalez D, Chinesta F, Cueto E (2021) Deep learning of thermodynamics-aware reduced-order models from data. Comput. Methods Appl. Mech. Eng. 379:113763
Finn C, Levine S (2017) Deep visual foresight for planning robot motion. In: 2017 IEEE International Conference on Robotics and Automation (ICRA), pp 2786–2793. IEEE
Liu CK, Negrut D (2021) The role of physics-based simulators in robotics. Ann Rev Control Robot Autonom Syst 4:35–58
Allen KR, Smith KA, Tenenbaum JB (2020) Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning. Proc Natl Acad Sci 117(47):29302–29310
Battaglia PW, Hamrick JB, Bapst V, Sanchez-Gonzalez A, Zambaldi V, Malinowski M, Tacchetti A, Raposo D, Santoro A, Faulkner R, et al (2018) Relational inductive biases, deep learning, and graph networks. arXiv preprint arXiv:1806.01261
Zheng N, Liu Z, Ren P, Ma Y, Chen S, Yu S, Xue J, Chen B, Wang F (2017) Hybrid-augmented intelligence: collaboration and cognition. Front Inf Technol Electron Eng 18(2):153–179
Rui Y (2017) From artificial intelligence to augmented intelligence. IEEE MultiMed 24(1):4–5
Chinesta F, Cueto E, Abisset-Chavanne E, Duval JL, Khaldi FE (2020) Virtual, digital and hybrid twins: a new paradigm in data-based engineering and engineered data. Arch Comput Methods Eng 27:105–134
Moya B, Badías A, Alfaro I, Chinesta F, Cueto E (2022) Digital twins that learn and correct themselves. Int J Numer Methods Eng 123(13):3034–3044
Cerf VG (2013) Augmented intelligence. IEEE Internet Comput 17(5):96–96
Bar M (2009) The proactive brain: memory for predictions. Philos Trans R Soc B 364(1521):1235–1243
Hamrick J, Battaglia P, Tenenbaum JB (2011) Internal physics models guide probabilistic judgments about object dynamics. In: Proceedings of the 33rd annual conference of the Cognitive Science Society, vol 2. Cognitive Science Society, Austin, TX
Battaglia PW, Hamrick JB, Tenenbaum JB (2013) Simulation as an engine of physical scene understanding. Proc Natl Acad Sci 110(45):18327–18332
Traylor A, Feiman R, Pavlick E (2022) Can neural networks learn implicit logic from physical reasoning? In: The eleventh international conference on learning representations
Murphy KP, Torralba A, Freeman W (2003) Using the forest to see the trees: a graphical model relating features, objects, and scenes. In: Advances in neural information processing systems 16
Gupta A, Efros AA, Hebert M (2010) Blocks world revisited: Image understanding using qualitative geometry and mechanics. In: Computer Vision–ECCV 2010: 11th European conference on computer vision, Heraklion, Crete, Greece, September 5–D11, 2010, Proceedings, Part IV 11, pp 482–496. Springer
Schenck C, Fox D (2016) Detection and tracking of liquids with fully convolutional networks. arXiv preprint arXiv:1606.06266
Shen B, Yan X, Qi CR, Najibi M, Deng B, Guibas L, Zhou Y, Anguelov D (2023) Gina-3d: learning to generate implicit neural assets in the wild. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 4913–4926
Kandukuri R, Achterhold J, Moeller M, Stueckler J (2020) Learning to identify physical parameters from video using differentiable physics. In: DAGM German conference on pattern recognition, pp 44–57. Springer
Wu J, Yildirim I, Lim JJ, Freeman B, Tenenbaum J (2015) Galileo: Perceiving physical object properties by integrating a physics engine with deep learning. In: Advances in neural information processing systems 28
Mrowca D, Zhuang C, Wang E, Haber N, Fei-Fei LF, Tenenbaum J, Yamins DL (2018) Flexible neural representation for physics prediction. In: Advances in neural information processing systems 31
Bender J, Erleben K, Trinkle J (2014) Interactive simulation of rigid body dynamics in computer graphics. In: Computer graphics forum, vol. 33, pp 246–270. Wiley Online Library
Rath L, Geist AR, Trimpe S (2022) Using physics knowledge for learning rigid-body forward dynamics with gaussian process force priors. In: Conference on robot learning, pp 101–111. PMLR
Huang S, Cheng Z-Q, Li X, Wu X, Zhang Z, Hauptmann A (2018) Perceiving physical equation by observing visual scenarios. arXiv preprint arXiv:1811.12238
Li Y, Wu J, Tedrake R, Tenenbaum JB, Torralba A (2018) Learning particle dynamics for manipulating rigid bodies, deformable objects, and fluids. arXiv preprint arXiv:1810.01566
Schenck C, Fox D (2018) Spnets: differentiable fluid dynamics for deep neural networks. In: Conference on robot learning, pp 317–335. PMLR
Duchaine V, Gosselin C (2009) Safe, stable and intuitive control for physical human-robot interaction. In: 2009 IEEE international conference on robotics and automation, pp 3383–3388. IEEE
Koppula HS, Saxena A (2015) Anticipating human activities using object affordances for reactive robotic response. IEEE Trans Pattern Anal Mach Intell 38(1):14–29
Liu X-Y, Wang J-X (2021) Physics-informed dyna-style model-based deep reinforcement learning for dynamic control. Proc R Soc A 477(2255):20210618
Driess D, Schubert I, Florence P, Li Y, Toussaint M (2022) Reinforcement learning with neural radiance fields. Adv Neural Inf Process Syst 35:16931–16945
Badias A, Alfaro I, Gonzalez D, Chinesta F, Cueto E (2021) Morph-dslam: model order reduction for physics-based deformable slam. IEEE Trans Pattern Anal Mach Intell 44(11):7764–7777
Wei B, Zhao Y, Hao K, Gao L (2021) Visual sensation and perception computational models for deep learning: state of the art, challenges and prospects. arXiv preprint arXiv:2109.03391
Assen JJR, Nishida S, Fleming RW (2020) Visual perception of liquids: insights from deep neural networks. PLoS Comput Biol 16(8):1008018
Zhang Y, Dong Z, Obaidat MS, Ban X (2023) Non-Newtonian fluid simulation and reconstruction from monocular videos. Simul Model Pract Theory 123:102688
Lopez-Guevara T, Pucci R, Taylor NK, Gutmann MU, Ramamoorthy S, Suhr K (2020) Stir to pour: Efficient calibration of liquid properties for pouring actions. In: 2020 IEEE/RSJ international conference on intelligent robots and systems (IROS), pp 5351–5357. IEEE
Della Santina C, Truby RL, Rus D (2020) Data-driven disturbance observers for estimating external forces on soft robots. IEEE Robot Autom Lett 5(4):5717–5724
Richter F, Orosco RK, Yip MC (2022) Image based reconstruction of liquids from 2d surface detections. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 13811–13820
Schenck C, Fox D (2018) Perceiving and reasoning about liquids using fully convolutional networks. Int J Robot Res 37(4–5):452–471
Kloss A, Schaal S, Bohg J (2022) Combining learned and analytical models for predicting action effects from sensory data. Int J Robot Res 41(8):778–797
Degrave J, Hermans M, Dambre J et al (2019) A differentiable physics engine for deep learning in robotics. Front Neurorobot 6:1
Avila Belbute-Peres F, Smith K, Allen K, Tenenbaum J, Kolter JZ (2018) End-to-end differentiable physics for learning and control. In: Advances in neural information processing systems 31
Ding M, Chen Z, Du T, Luo P, Tenenbaum J, Gan C (2021) Dynamic visual reasoning by learning differentiable physics models from video and language. Adv Neural Inf Process Syst 34:887–899
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. In: Advances in neural information processing systems 30
Rao C, Sun H, Liu Y (2021) Physics-informed deep learning for computational elastodynamics without labeled data. J Eng Mech 147(8):04021043
Callaham JL, Maeda K, Brunton SL (2019) Robust flow reconstruction from limited measurements via sparse representation. Phys Rev Fluids 4(10):103907
Yang T-Y, Rosca J, Narasimhan K, Ramadge PJ (2022) Learning physics constrained dynamics using autoencoders. Adv Neural Inf Process Syst 35:17157–17172
Marisca I, Cini A, Alippi C (2022) Learning to reconstruct missing data from spatiotemporal graphs with sparse observations. Adv Neural Inf Process Syst 35:32069–32082
Sun L, Wang J-X (2020) Physics-constrained Bayesian neural network for fluid flow reconstruction with sparse and noisy data. Theoret Appl Mech Lett 10(3):161–169
Tong Z, Li Y (2020) Real-time reconstruction of contaminant dispersion from sparse sensor observations with gappy pod method. Energies 13(8):1956
Li T, Buzzicotti M, Biferale L, Bonaccorso F, Chen S, Wan M (2022) Data reconstruction of turbulent flows with gappy pod, extended pod and generative adversarial networks. arXiv preprint arXiv:2210.11921
Mainini L, Willcox K (2015) Surrogate modeling approach to support real-time structural assessment and decision making. AIAA J 53(6):1612–1626
Demo N, Tezzele M, Rozza G (2023) A deeponet multi-fidelity approach for residual learning in reduced order modeling. arXiv preprint arXiv:2302.12682
Salam T, Hsieh MA (2019) Adaptive sampling and reduced-order modeling of dynamic processes by robot teams. IEEE Robot Automat Lett 4(2):477–484
Rovina H, Salam T, Kantaros Y, Hsieh MA (2020) Asynchronous adaptive sampling and reduced-order modeling of dynamic processes by robot teams via intermittently connected networks. In: 2020 IEEE/RSJ international conference on Intelligent Robots and Systems (IROS), pp 4798–4805. IEEE
Ebert C, Ruwisch C, Weiss J, Uijt De Haag M, Silvestre F (2022) Trajectory planning in windy urban environment–a gappy pod approach for wind field estimates with sparse sensors. In: AIAA AVIATION 2022 Forum, p 3757
Raben SG, Charonko JJ, Vlachos PP (2012) Adaptive gappy proper orthogonal decomposition for particle image velocimetry data reconstruction. Meas Sci Technol 23(2):025303
Kelshaw D, Rigas G, Magri L (2022) Physics-informed cnns for super-resolution of sparse observations on dynamical systems. arXiv preprint arXiv:2210.17319
Yu L, Yousif MZ, Zhang M, Hoyas S, Vinuesa R, Lim H-C (2022) Three-dimensional esrgan for super-resolution reconstruction of turbulent flows with tricubic interpolation-based transfer learning. Phys Fluids 34:12
Vinuesa R, Brunton SL, McKeon BJ (2023) The transformative potential of machine learning for experiments in fluid mechanics. arXiv preprint arXiv:2303.15832
Gao H, Sun L, Wang J-X (2021) Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels. Phys Fluids 33:7
t Wang X, Xie L, Dong C, Shan Y (2021) Real-esrgan: Training real-world blind super-resolution with pure synthetic data. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 1905–1914
Chen H, He X, Qing L, Wu Y, Ren C, Sheriff RE, Zhu C (2022) Real-world single image super-resolution: a brief review. Inf Fusion 79:124–145
Saharia C, Ho J, Chan W, Salimans T, Fleet DJ, Norouzi M (2022) Image super-resolution via iterative refinement. IEEE Trans Pattern Anal Mach Intell 45(4):4713–4726
Nair A, Chen D, Agrawal P, Isola P, Abbeel P, Malik J, Levine S (2017) Combining self-supervised learning and imitation for vision-based rope manipulation. In: 2017 IEEE International Conference on Robotics and Automation (ICRA), pp 2146–2153. IEEE
Nava M, Paolillo A, Guzzi J, Gambardella LM, Giusti A (2021) Uncertainty-aware self-supervised learning of spatial perception tasks. IEEE Robot Automat Lett 6(4):6693–6700
Yan M, Zhu Y, Jin N, Bohg J (2020) Self-supervised learning of state estimation for manipulating deformable linear objects. IEEE Robot Automat Lett 5(2):2372–2379
Flaschel M, Kumar S, De Lorenzis L (2021) Unsupervised discovery of interpretable hyperelastic constitutive laws. Comput Methods Appl Mech Eng 381:113852
Lye KO, Mishra S, Ray D (2020) Deep learning observables in computational fluid dynamics. J Comput Phys 410:109339
Shin Y-S, Kim J (2023) Sensor data reconstruction for dynamic responses of structures using external feedback of recurrent neural network. Sensors 23(5):2737
Moya B, Badias A, Gonzalez D, Chinesta F, Cueto E (2022) Physics perception in sloshing scenes with guaranteed thermodynamic consistency. IEEE Trans Pattern Anal Mach Intell 45(2):2136–2150
Sun C, Karlsson P, Wu J, Tenenbaum JB (2019) Murphy, K.: Stochastic prediction of multi-agent interactions from partial observations. arXiv preprint arXiv:1902.09641
Antonova R, Yang J, Sundaresan P, Fox D, Ramos F, Bohg J (2022) A Bayesian treatment of real-to-sim for deformable object manipulation. IEEE Robot Automat Lett 7(3):5819–5826
Lim V, Huang H, Chen LY, Wang J, Ichnowski J, Seita D, Laskey M, Goldberg K (2022) Real2sim2real: self-supervised learning of physical single-step dynamic actions for planar robot casting. In: 2022 International Conference on Robotics and Automation (ICRA), pp 8282–8289. IEEE
Li Y, Torralba A, Anandkumar A, Fox D, Garg A (2020) Causal discovery in physical systems from videos. Adv Neural Inf Process Syst 33:9180–9192
Bai Z, Brunton SL, Brunton BW, Kutz JN, Kaiser E, Spohn A (2017) Noack, B.R.: Data-driven methods in fluid dynamics: sparse classification from experimental data. Springer, Berlin
Rodríguez-Ocampo P, Ring M, Hernández-Fontes J, Alcérreca-Huerta J, Mendoza E, Gallegos-Diez-Barroso G, Silva R (2020) A 2d image-based approach for cfd validation of liquid mixing in a free-surface condition. J Appl Fluid Mech 13(5):1487–1500
Bieker K, Peitz S, Brunton SL, Kutz JN, Dellnitz M (2020) Deep model predictive flow control with limited sensor data and online learning. Theoret Comput Fluid Dyn 34:577–591
Thuruthel TG, Shih B, Laschi C, Tolley MT (2019) Soft robot perception using embedded soft sensors and recurrent neural networks. Sci Robot 4(26):1488
Tariverdi A, Venkiteswaran VK, Richter M, Elle OJ, Tørresen J, Mathiassen K, Misra S, Martinsen ØG (2021) A recurrent neural-network-based real-time dynamic model for soft continuum manipulators. Front Robot AI 8:631303
Bonassi F, Farina M, Xie J, Scattolini R (2022) On recurrent neural networks for learning-based control: recent results and ideas for future developments. J Process Control 114:92–104
Ehrhardt S, Monszpart A, Mitra NJ, Vedaldi A (2019) Taking visual motion prediction to new heightfields. Comput Vis Image Understand 181:14–25
Buschoff LMS, Schulz E, Binz M (2023) The acquisition of physical knowledge in generative neural networks
Doerr A, Daniel C, Schiegg M, Duy N-T, Schaal S, Toussaint M, Sebastian T (2018) Probabilistic recurrent state-space models. In: International conference on machine learning, pp 1280–1289. PMLR
Rai R, Sahu CK (2020) Driven by data or derived through physics? a review of hybrid physics guided machine learning techniques with cyber-physical system (cps) focus. IEEE Access 8:71050–71073
Achterhold J, Tobuschat P, Ma H, Büchler D, Muehlebach M, Stueckler J (2023) Black-box vs. gray-box: A case study on learning table tennis ball trajectory prediction with spin and impacts. In: Learning for Dynamics and Control Conference (L4DC). accepted
Driess D, Huang Z, Li Y, Tedrake R, Toussaint M (2023) Learning multi-object dynamics with compositional neural radiance fields. In: Conference on robot learning, pp 1755–1768. PMLR
Badías A, González D, Alfaro I, Chinesta F, Cueto E (2020) Real-time interaction of virtual and physical objects in mixed reality applications. Int J Numer Methods Eng 121(17):3849–3868
Zhong YD, Han J, Dey B, Brikis GO (2023) Improving gradient computation for differentiable physics simulation with contacts. In: Learning for dynamics and control conference, pp 128–141. PMLR
Jiang Y, Sun J, Liu CK (2022) Data-augmented contact model for rigid body simulation. In: Learning for dynamics and control conference, pp 378–390. PMLR
Strecke M, Stueckler J (2021) Diffsdfsim: differentiable rigid-body dynamics with implicit shapes. In: 2021 international conference on 3D Vision (3DV), pp 96–105. IEEE
Hernández Q, Badías A, Chinesta F, Cueto E (2023) Port-metriplectic neural networks: thermodynamics-informed machine learning of complex physical systems. Comput Mech 1:1–9
Um K, Brand R, Fei YR, Holl P, Thuerey N (2020) Solver-in-the-loop: Learning from differentiable physics to interact with iterative pde-solvers. Adv Neural Inf Process Syst 33:6111–6122
Wiewel S, Kim B, Azevedo VC, Solenthaler B, Thuerey N (2020) Latent space subdivision: stable and controllable time predictions for fluid flow. In: Computer graphics forum, vol. 39, pp 15–25. Wiley Online Library
Takahashi T, Liang J, Qiao Y-L, Lin MC (2021) Differentiable fluids with solid coupling for learning and control. In: Proceedings of the AAAI conference on artificial intelligence, vol. 35, pp 6138–6146
Karniadakis GE, Kevrekidis IG, Lu L, Perdikaris P, Wang S, Yang L (2021) Physics-informed machine learning. Nat Rev Phys 3(6):422–440
Ayensa-Jimenez J, Doweidar MH, Sanz-Herrera JA, Doblare M (2021) Prediction and identification of physical systems by means of physically-guided neural networks with meaningful internal layers. Comput Methods Appl Mech Eng 381:113816
Ajay A, Wu J, Fazeli N, Bauza M, Kaelbling LP, Tenenbaum JB, Rodriguez A (2018) Augmenting physical simulators with stochastic neural networks: Case study of planar pushing and bouncing. In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp 3066–3073. IEEE
Zeng A, Song S, Lee J, Rodriguez A, Funkhouser T (2020) Tossingbot: learning to throw arbitrary objects with residual physics. IEEE Trans Robot 36(4):1307–1319
Allevato A, Pryor M, Thomaz AL (2021) Multiparameter real-world system identification using iterative residual tuning. J Mech Robot 13(3):031021
Rudin C, Chen C, Chen Z, Huang H, Semenova L, Zhong C (2022) Interpretable machine learning: fundamental principles and 10 grand challenges. Stat Surv 16:1–85
Lu L, Pestourie R, Yao W, Wang Z, Verdugo F, Johnson SG (2021) Physics-informed neural networks with hard constraints for inverse design. SIAM J Sci Comput 43(6):1105–1132
Liu S, Zhongkai H, Ying C, Su H, Zhu J, Cheng Z (2022) A unified hard-constraint framework for solving geometrically complex pdes. Adv Neural Inf Process Syst 35:20287–20299
Alkhadhr S, Almekkawy M (2023) Wave equation modeling via physics-informed neural networks: models of soft and hard constraints for initial and boundary conditions. Sensors 23(5):2792
Wu W, Daneker M, Jolley MA, Turner KT, Lu L (2023) Effective data sampling strategies and boundary condition constraints of physics-informed neural networks for identifying material properties in solid mechanics. Appl Math Mech 44(7):1039–1068
Prantl L, Ummenhofer B, Koltun V, Thuerey N (2022) Guaranteed conservation of momentum for learning particle-based fluid dynamics. Adv Neural Inf Process Syst 35:6901–6913
Tang J, Kim B, Azevedo VC, Solenthaler B (2023) Physics-informed neural corrector for deformation-based fluid control. In: Computer Graphics Forum, vol. 42, pp 161–173. Wiley Online Library
Raissi M, Perdikaris P, Karniadakis GE (2019) Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J Comput Phys 378:686–707
Huang Z, Sun Y, Wang W (2023) Generalizing graph ode for learning complex system dynamics across environments. arXiv preprint arXiv:2307.04287
Huang Z, Sun Y, Wang W (2021) Coupled graph ode for learning interacting system dynamics. In: Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, pp 705–715
Huang Z, Sun Y, Wang W (2020) Learning continuous system dynamics from irregularly-sampled partial observations. Adv Neural Inf Process Syst 33:16177–16187
Weinan E (2017) A proposal on machine learning via dynamical systems. Commun Math Stat 1(5):1–11
Wen G, Li D, Qin F (2022) Learning symplectic dynamics via generating recurrent neural network. In: Proceedings of the 2022 5th international conference on machine learning and machine intelligence, pp 65–70
Jin P, Zhang Z, Zhu A, Tang Y, Karniadakis GE (2020) Sympnets: intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems. Neural Netw 132:166–179
Chen Z, Feng M, Yan J, Zha H (2022) Learning neural Hamiltonian dynamics: a methodological overview. arXiv preprint arXiv:2203.00128
Cranmer M, Greydanus S, Hoyer S, Battaglia P, Spergel D, Ho S (2020) Lagrangian neural networks. arXiv preprint arXiv:2003.04630
Trask N, Huang A, Hu X (2022) Enforcing exact physics in scientific machine learning: a data-driven exterior calculus on graphs. J Comput Phys 456:110969
Patel RG, Manickam I, Trask NA, Wood MA, Lee M, Tomas I, Cyr EC (2022) Thermodynamically consistent physics-informed neural networks for hyperbolic systems. J Comput Phys 449:110754
Vlassis NN, Sun W (2021) Sobolev training of thermodynamic-informed neural networks for interpretable elasto-plasticity models with level set hardening. Comput Methods Appl Mech Eng 377:113695
Kaltenbach S, Koutsourelakis P-S (2020) Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems. J Comput Phys 419:109673
Masi F, Stefanou I, Vannucci P, Maffi-Berthier V (2021) Thermodynamics-based artificial neural networks for constitutive modeling. J Mech Phys Solids 147:104277
He X, Chen J-S (2022) Thermodynamically consistent machine-learned internal state variable approach for data-driven modeling of path-dependent materials. Comput Methods Appl Mech Eng 402:115348
Yu H, Tian X, Weinan E, Li Q (2021) Onsagernet: learning stable and interpretable dynamics using a generalized onsager principle. Phys Rev Fluids 6(11):114402
Huang S, He Z, Reina C (2022) Variational onsager neural networks (vonns): a thermodynamics-based variational learning strategy for non-equilibrium pdes. J Mech Phys Solids 163:104856
Biot M (1955) Variational principles in irreversible thermodynamics with application to viscoelasticity. Phys Rev 97(6):1463
Morrison PJ (1986) A paradigm for joined Hamiltonian and dissipative systems. Physica D 18(1–3):410–419
Öttinger HC, Grmela M (1997) Dynamics and thermodynamics of complex fluids. ii. Illustrations of a general formalism. Phys Rev E 56(6):6633
Lee K, Trask N, Stinis P (2021) Machine learning structure preserving brackets for forecasting irreversible processes. Adv Neural Inf Process Syst 34:5696–5707
Zhang Z, Shin Y, Em Karniadakis G (2022) Gfinns: generic formalism informed neural networks for deterministic and stochastic dynamical systems. Philos Trans R Soc A 380(2229):20210207
Hernández Q, Badías A, González D, Chinesta F, Cueto E (2021) Structure-preserving neural networks. J Comput Phys 426:109950
Hernández Q, Badías A, Chinesta F, Cueto E (2023) Thermodynamics-informed neural networks for physically realistic mixed reality. Comput Methods Appl Mech Eng 407:115912
Hernández Q, Badías A, Chinesta F, Cueto E (2022) Thermodynamics-informed graph neural networks. arXiv preprint arXiv:2203.01874
Liu J, Shen H, Wang D, Kang Y, Tian Q (2021) Unsupervised domain adaptation with dynamics-aware rewards in reinforcement learning. Adv Neural Inf Process Syst 34:28784–28797
Liu J, Zhang H, Zhuang Z, Kang Y, Wang D, Wang B (2023) Design from policies: conservative test-time adaptation for offline policy optimization. arXiv preprint arXiv:2306.14479
Short ES, Allevato A, Thomaz AL (2019) Sail: simulation-informed active in-the-wild learning. In: 2019 14th ACM/IEEE international conference on Human-Robot Interaction (HRI), pp 468–477. IEEE
Allevato A, Short ES, Pryor M, Thomaz A (2020) Tunenet: one-shot residual tuning for system identification and sim-to-real robot task transfer. In: Conference on robot learning, pp 445–455. PMLR
Lutter M, Peters J (2023) Combining physics and deep learning to learn continuous-time dynamics models. Int J Robot Res 42(3):83–107
Hussein A, Gaber MM, Elyan E, Jayne C (2017) Imitation learning: a survey of learning methods. ACM Comput Surv 50(2):1–35
Ho J, Ermon S (2016) Generative adversarial imitation learning. Adv Neural Inf Process Syst 29
Chen S, Ma X, Xu Z (2023) Imitation learning as state matching via differentiable physics. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 7846–7855
Adams M, Li X, Boucinha L, Kher SS, Banerjee P, Gonzalez J-L (2021) Hybrid digital twins: a primer on combining physics-based and data analytics approaches. IEEE Softw 39(2):47–52
Sancarlos A, Cameron M, Abel A, Cueto E, Duval J-L, Chinesta F (2021) From rom of electrochemistry to ai-based battery digital and hybrid twin. Arch Comput Methods Eng 28:979–1015
Desai S, Durugkar I, Karnan H, Warnell G, Hanna J, Stone P (2020) An imitation from observation approach to transfer learning with dynamics mismatch. Adv Neural Inf Process Syst 33:3917–3929
Mohebujjaman M, Rebholz LG, Iliescu T (2019) Physically constrained data-driven correction for reduced-order modeling of fluid flows. Int J Numer Methods Fluids 89(3):103–122
Goswami S, Anitescu C, Chakraborty S, Rabczuk T (2020) Transfer learning enhanced physics informed neural network for phase-field modeling of fracture. Theoret Appl Fract Mech 106:102447
Laroche R, Barlier M (2017) Transfer reinforcement learning with shared dynamics. In: Proceedings of the AAAI conference on artificial intelligence, vol. 31
Guastoni L, Güemes A, Ianiro A, Discetti S, Schlatter P, Azizpour H, Vinuesa R (2021) Convolutional-network models to predict wall-bounded turbulence from wall quantities. J Fluid Mech 928:27
Moya B, Badías A, González D, Chinesta F, Cueto E (2023) A thermodynamics-informed active learning approach to perception and reasoning about fluids. Comput Mech 1:1–15
Monaghan JJ (1992) Smoothed particle hydrodynamics. Ann Rev Astron Astrophys 30(1):543–574
Bates C, Battaglia PW, Yildirim I, Tenenbaum JB (20185) Humans predict liquid dynamics using probabilistic simulation. In: CogSci
Espanol P, Serrano M, Öttinger HC (1999) Thermodynamically admissible form for discrete hydrodynamics. Phys Rev Lett 83(22):4542
Moya B, Alfaro I, Gonzalez D, Chinesta F, Cueto E (2020) Physically sound, self-learning digital twins for sloshing fluids. PLoS ONE 15(6):0234569
Brunton SL, Proctor JL, Kutz JN (2016) Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc Natl Acad Sci 113(15):3932–3937
Pascanu R, Mikolov T, Bengio Y (2013) On the difficulty of training recurrent neural networks. In: International conference on machine learning, pp 1310–1318. Pmlr
Chung J, Gulcehre C, Cho K, Bengio Y (2014) Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555
Kubo R (1966) The fluctuation-dissipation theorem. Rep Prog Phys 29(1):255
Kraus M (2021) Metriplectic integrators for dissipative fluids. In: Geometric science of information: 5th international conference, GSI 2021, Paris, France, July 21–23, 2021, Proceedings 5, pp 292–301. Springer
Grmela M, Öttinger HC (1997) Dynamics and thermodynamics of complex fluids. i. Development of a general formalism. Phys Rev E 56(6):6620
Reyes B, Howard AA, Perdikaris P, Tartakovsky AM (2021) Learning unknown physics of non-Newtonian fluids. Phys Rev Fluids 6(7):073301
Cranmer M, Sanchez Gonzalez A, Battaglia P, Xu R, Cranmer K, Spergel D, Ho S (2020) Discovering symbolic models from deep learning with inductive biases. Adv Neural Inf Process Syst 33:17429–17442
Arrieta AB, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A, García S, Gil-López S, Molina D, Benjamins R (2020) Explainable artificial intelligence (xai): concepts, taxonomies, opportunities and challenges toward responsible ai. Inf fusion 58:82–115
Hao Z, Liu S, Zhang Y, Ying C, Feng Y, Su H, Zhu J (2022) Physics-informed machine learning: a survey on problems, methods and applications. arXiv preprint arXiv:2211.08064
Kadambi A, Melo C, Hsieh C-J, Srivastava M, Soatto S (2023) Incorporating physics into data-driven computer vision. Nat Mach Intell 1:1–9
Banerjee C, Nguyen K, Fookes C, Karniadakis G (2023) Physics-informed computer vision: a review and perspectives. arXiv preprint arXiv:2305.18035
LeCun Y (2022) A path towards autonomous machine intelligence version 0.9. 2, 2022-06-27. Open Review 62
Silver D, Singh S, Precup D, Sutton RS (2021) Reward is enough. Artif Intell 299:103535
Acknowledgements
This research is part of the DesCartes program and is supported by the National Research Foundation, Prime Minister Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme. This material is also based upon work supported in part by the Army Research Laboratory and the Army Research Office under contract/Grant No. W911NF2210271. This work has been partially funded by the Spanish Ministry of Science and Innovation, AEI /10.13039/501100011033, through Grant No. TED2021-130105B-I00. The authors also acknowledge the support of ESI Group through the chairs at the University of Zaragoza and at ENSAM Institute of Technology.
Funding
This research has been supported by the Army Research Laboratory and the Army Research Office under contract/grant number W911NF2210271, the Spanish Ministry of Science and Innovation, AEI /10.13039/501100011033, through Grant No. TED2021-130105B-I00, ESI Group through the chairs at the University of Zaragoza and at ENSAM Institute of Technology and he National Research Foundation, Prime Minister Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme.
Author information
Authors and Affiliations
Contributions
All authors declare they have contributed equally to this work and there is no interest conflict.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no financial or non-financial interests to disclose.
Ethical Approval
Not applicable.
Consent for Publication
All authors consent the publication of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Moya, B., Badías, A., González, D. et al. Computational Sensing, Understanding, and Reasoning: An Artificial Intelligence Approach to Physics-Informed World Modeling. Arch Computat Methods Eng 31, 1897–1914 (2024). https://doi.org/10.1007/s11831-023-10033-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11831-023-10033-y