mlsanityData to 14 Oct 2026

Discovering physical concepts with neural networks

arXiv 1807.10300 ↗First posted 26 Jul 2018quant-ph cs.LG physics.data-ancited by 117

Not on the map. Why?

Abstract (arXiv)

Despite the success of neural networks at solving concrete physics problems, their use as a general-purpose tool for scientific discovery is still in its infancy. Here, we approach this problem by modelling a neural network architecture after the human physical reasoning process, which has similarities to representation learning. This allows us to make progress towards the long-term goal of machine-assisted scientific discovery from experimental data without making prior assumptions about the system. We apply this method to toy examples and show that the network finds the physically relevant parameters, exploits conservation laws to make predictions, and can help to gain conceptual insights, e.g. Copernicus' conclusion that the solar system is heliocentric.

Key facts. Cited by 117 papers in the map (data through 14 Oct 2026). First posted on arXiv 26 Jul 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 29; 2022: 21; 2023: 25; 2024: 13; 2025: 17; 2026: 12.

117 papers.

Sort:

By area:

  1. 2022SELFIES and the future of molecular string representationscited by 76
  2. 2022On scientific understanding with artificial intelligencecited by 70
  3. 2025What Has a Foundation Model Found? Using Inductive Bias to Probe for World Modelscited by 47
  4. 2021Towards extraction of orthogonal and parsimonious non-linear modes from turbulent flowscited by 45
  5. 2021Power-law escape rate of SGDcited by 29
  6. 2021Towards a Collective Agenda on AI for Earth Science Data Analysiscited by 27
  7. 2021Machine-Learning Mathematical Structurescited by 25
  8. 2023Deep Learning Symmetries and Their Lie Groups, Algebras, and Subalgebras from First Principlescited by 22
  9. 2021Adaptable Hamiltonian neural networkscited by 21
  10. 2022Neural Implicit Representations for Physical Parameter Inference from a Single Videocited by 19
  11. 2022Improving aircraft performance using machine learning: a reviewcited by 19
  12. 2023Evolving Scientific Discovery by Unifying Data and Background Knowledge with AI Hilbertcited by 18
38 more
  1. 2021MC-LSTM: Mass-Conserving LSTMcited by 17
  2. 2021Solving time dependent Fokker-Planck equations via temporal normalizing flowcited by 17
  3. 2023Latent Traversals in Generative Models as Potential Flowscited by 17
  4. 2022Flexible learning of quantum states with generative query neural networkscited by 16
  5. 2025Interpretable Machine Learning in Physics: A Reviewcited by 14
  6. 2024Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systemscited by 13
  7. 2021Learning Interpretable Representations of Entanglement in Quantum Optics Experiments using Deep Generative Modelscited by 12
  8. 2022Quantum Similarity Testing with Convolutional Neural Networkscited by 10
  9. 2023Explainable Representation Learning of Small Quantum Statescited by 10
  10. 2024Generalized Lagrangian Neural Networkscited by 10
  11. 2025Artificial intelligence for representing and characterizing quantum systemscited by 10
  12. 2023Metalearning generalizable dynamics from trajectoriescited by 9
  13. 2023Physics-informed Information Field Theory for Modeling Physical Systems with Uncertainty Quantificationcited by 9
  14. 2023Symbolic Regression via Control Variable Genetic Programmingcited by 9
  15. 2021Discovering conservation laws from trajectories via machine learningcited by 8
  16. 2021Disentangled Generative Models for Robust Prediction of System Dynamicscited by 8
  17. 2021AI Descartes: Combining Data and Theory for Derivable Scientific Discoverycited by 8
  18. 2021Accelerating Understanding of Scientific Experiments with End to End Symbolic Regressioncited by 8
  19. 2022Is the Machine Smarter than the Theorist: Deriving Formulas for Particle Kinematics with Symbolic Regressioncited by 8
  20. 2023Discovering Sparse Representations of Lie Groups with Machine Learningcited by 8
  21. 2022A Review of Machine Learning Methods Applied to Structural Dynamics and Vibroacousticcited by 7
  22. 2023Learning minimal representations of stochastic processes with variational autoencoderscited by 7
  23. 2024Learning Hamiltonian neural Koopman operator and simultaneously sustaining and discovering conservation lawcited by 7
  24. 2024Creativity in AI: Progresses and Challengescited by 7
  25. 2025Learning mechanical systems from real-world data using discrete forced Lagrangian dynamicscited by 7
  26. 2025Inverse Design in Nanophotonics via Representation Learningcited by 7
  27. 2025The Need for Verification in AI-Driven Scientific Discoverycited by 7
  28. 2021Manifold Learning for Knowledge Discovery and Intelligent Inverse Design of Photonic Nanostructures: Breaking the Geometric Complexitycited by 6
  29. 2023Accelerated Discovery of Machine-Learned Symmetries: Deriving the Exceptional Lie Groups G2, F4 and E6cited by 6
  30. 2023Identifying the Group-Theoretic Structure of Machine-Learned Symmetriescited by 6
  31. 2024Vertical Symbolic Regression via Deep Policy Gradientcited by 6
  32. 2024Oil & Water? Diffusion of AI Within and Across Scientific Fieldscited by 6
  33. 2025Quantum computing and artificial intelligence: status and perspectivescited by 6
  34. 2025Interpretable representation learning of quantum data enabled by probabilistic variational autoencoderscited by 6
  35. 2022Discovering the building blocks of dark matter halo density profiles with neural networkscited by 5
  36. 2023Oracle-Preserving Latent Flowscited by 5
  37. 2023Deep Quantum Graph Dreaming: Deciphering Neural Network Insights into Quantum Experimentscited by 5
  38. 2025AI-Newton: A Concept-Driven Physical Law Discovery System without Prior Physical Knowledgecited by 5

50 most cited of 117 All 117 (JSON)