Discovering physical concepts with neural networks

By Raban Iten, Tony Metger, Henrik Wilming, Lídia del Rio, Renato Renner. 2018

447 citations recorded by the source corpus.

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 modeling 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.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistical and Nonlinear Physics

Citations in the corpus

Weakly-Supervised Disentanglement Without Compromises (2020, background): “This result is of great importance as the community is becoming increasingly interested in the practical benefits of disentangled representations (van Steenkiste et al., 2019; Locatello et al., 2019a; Creager et al., 2019; Chao et al., 2019; Iten et al., 2020; Chartsias et al., 2019; Higgins et al.,”
Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation (2019, background): “n excellent predictive or generative results. In the best case, they understand what the model has learned – interpret its result [43, 44, 45], in particular interpreting its internal representations [46]. This task seems highly challenging when most of the latent space leads to invalid, chemical or”
Deep Learning for Geophysics: Current and Future Trends (2021, methodology): “DL can even be used for discovering physical concepts (Iten et al., 2020), such as the solar system is heliocentric.”
Explainable Machine Learning for Scientific Insights and Discoveries (2019, background): “Reference [100] introduces SciNet, a modified variational autoencoder that learns a representation from experimental data and uses the learned representation to derive physical concepts from it rather than from the experimental input”
Hamiltonian Neural Networks (2019, background): “Two recent works are similar to this paper in that the authors sought to uncover physical laws from data using neural networks [18, 4].”
On scientific understanding with artificial intelligence (2022, methodology): “[21] Raban Iten, Tony Metger, Henrik Wilming, Ĺıdia”