Absence of hidden analytic conserved quantities in harmonically confined rods
arXiv:2607.18872
2026
Architecture
1 ideas extracted · analyzed Aug 30, 2026
What the math gives to ML
The paper gives a constructive symmetry characterization of analytic conserved quantities in harmonically trapped hard-rod dynamics. Free harmonic motion imposes invariance under phase rotations, while finite-size elastic collisions impose permutation invariance on momenta; together these conditions imply that every analytic invariant is a function of total energy and center-of-mass energy. The transferable neural-network mechanism is a symmetry-constrained invariant module or critic for collision-rich world models, with a falsifiable test that learned conserved scalars contain no independent information beyond these two energies.
Ideas from this paper
Unverified
2026
Construct a scalar feature or critic for oscillator-based neural dynamics that is invariant under the transformations imposed by free harmonic motion and elastic collisions. For finite-size rods, the module should represent only quantities compatible with common oscillator-phase rotations and momentum permutations, preventing a learned world model from inventing coordinate-dependent pseudo-conserved quantities that disappear after collisions.
Useful6/10
Difficulty5/10
Novelty6/10