First poly-time algorithm for dihedral and projected MRA via recursive method of moments on the third moment tensor, conditional on a verifiable rank conjecture for power-of-two lengths.
Group-invariant moments under tomographic projections
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
Let $f:\mathbb{R}^n\to\mathbb{R}$ be an unknown object, and suppose the observations are tomographic projections of randomly rotated copies of $f$ of the form $Y = P(R\cdot f)$, where $R$ is Haar-uniform in $\mathrm{SO}(n)$ and $P$ is the projection onto an $m$-dimensional subspace, so that $Y:\mathbb{R}^m\to\mathbb{R}$. We prove that, whenever $d\le m$, the $d$-th order moment of the projected data determines the full $d$-th order Haar-orbit moment of $f$, independently of the ambient dimension $n$. We further provide an explicit algorithmic procedure for recovering the latter from the former. As a consequence, any identifiability result for the unprojected model based on the $d$-th order group-invariant moment extends directly to the tomographic setting at the same moment order. In particular, for $n=3$, $m=2$, and $d=2$, our result recovers a classical result in the cryo-EM literature: the covariance of the 2D projection images determines the second order rotationally invariant moment of the underlying 3D object.
years
2026 4representative citing papers
Develops functional multi-target detection theory and recovery algorithms via bispectrum inversion with non-asymptotic guarantees for compactly supported signals under continuous translations and correlated Gaussian noise.
The first three moments determine generic dihedral orbits in the projected MRA model under high noise via reduction to dihedral phase-coupling, with a constructive scheme and supporting experiments.
In low-SNR Gaussian latent-variable models, optimally weighted GMoM using minimal-order moments achieves the same leading asymptotic covariance as MLE via matching layerwise expansions of the information operators.
citing papers explorer
-
Functional Multi-Target Detection via Bispectrum Inversion
Develops functional multi-target detection theory and recovery algorithms via bispectrum inversion with non-asymptotic guarantees for compactly supported signals under continuous translations and correlated Gaussian noise.
-
Projected multi-reference alignment
The first three moments determine generic dihedral orbits in the projected MRA model under high noise via reduction to dihedral phase-coupling, with a constructive scheme and supporting experiments.