nLab physical neural network

 Idea

A neural network which relies on a physical medium, such as a nonlinear? optical system, a mechanical plate oscillator, or an electronic? circuit?, to emulate neurons, rather than software? code?.

See also

• neural network

• machine learning

 References

• Logan G. Wright, Tatsuhiro Onodera, Martin M. Stein, Tianyu Wang, Darren T. Schachter, Zoey Hu & Peter L. McMahon, Deep physical neural networks trained with backpropagation, Nature 601, 549–555 (2022). doi:10.1038/s41586-021-04223-6. (web)

• Victor Lopez-Pastor, Florian Marquardt, Self-learning Machines based on Hamiltonian Echo Backpropagation, (arXiv:2103.04992)

• Sam Dillavou, Menachem Stern, Andrea J. Liu, Douglas J. Durian, Demonstration of Decentralized, Physics-Driven Learning, (arXiv:2108.00275)