Kryptovaluta-ticker:
technology fra Arxiv cs.ai

Exact equivariance, kept through training, buys zero-shot generalisation across the symmetry group

Hongbo Wang (Stony Brook University)
Jun 3, 2026 at 04:00
9 Visninger
0 Kommentarer

arXiv:2606.03003v1 Announce Type: cross Abstract: A latent world model built from an equivariant encoder $E$ and an equivariant predictor $f$ inherits a provable symmetry of its training loss: when the world's dynamics genuinely carries a group $G$ acting on latents by an orthogonal representation $\rho(g)$, the one-step prediction relMSE is...

Les hele artikkelen hos kilden.

Var dette nyttig?
Del:

Kommentarer (0)

Vennligst logg inn for å skrive en kommentar

Ingen kommentarer ennå. Bli den første til å kommentere!