#physicsinformedml — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #physicsinformedml, aggregated by home.social.
-
I wrote up physics-informed ML on The Gradient: what it means, the methods (PINNs, gray-box, hard-constrained architectures), the three conditions where it wins, and when to just use a black box.
https://thegradient.io/physics-informed-machine-learning-when-it-beats-a-black-box
#MachineLearning #PhysicsInformedML #AI #Robotics #DeepLearning #PINN
-
Representation learning often emphasizes metric preservation. We instead build Symplectic structural invariance directly into the representation.
https://arxiv.org/abs/2512.19409
We embed Hamiltonian/symplectic geometry by making the RNN state dynamics a symplectomorphism, which preserves Legendre duality (information geometry) through time. This yields structure-preserving representations enforced by the latent dynamics, rather than imposed indirectly via the output.
#ReservoirComputing #RepresentationLearning #InformationGeometry #SymplecticGeometry #HamiltonianDynamics #GeometricDeepLearning #DynamicalSystems #PhysicsInformedML