Riemannian Optimization for Distance Geometry: A Study of Convergence, Robustness, and Incoherence
By: Chandler Smith, HanQin Cai, Abiy Tasissa
Potential Business Impact:
Finds hidden shapes from incomplete distance clues.
The problem of recovering a configuration of points from partial pairwise distances, referred to as the Euclidean Distance Geometry (EDG) problem, arises in a broad range of applications, including sensor network localization, molecular conformation, and manifold learning. In this paper, we propose a Riemannian optimization framework for solving the EDG problem by formulating it as a low-rank matrix completion task over the space of positive semi-definite Gram matrices. The available distance measurements are encoded as expansion coefficients in a non-orthogonal basis, and optimization over the Gram matrix implicitly enforces geometric consistency through the triangle inequality, a structure inherited from classical multidimensional scaling. Under a Bernoulli sampling model for observed distances, we prove that Riemannian gradient descent on the manifold of rank-$r$ matrices locally converges linearly with high probability when the sampling probability satisfies $p \geq \mathcal{O}(\nu^2 r^2 \log(n)/n)$, where $\nu$ is an EDG-specific incoherence parameter. Furthermore, we provide an initialization candidate using a one-step hard thresholding procedure that yields convergence, provided the sampling probability satisfies $p \geq \mathcal{O}(\nu r^{3/2} \log^{3/4}(n)/n^{1/4})$. A key technical contribution of this work is the analysis of a symmetric linear operator arising from a dual basis expansion in the non-orthogonal basis, which requires a novel application of the Hanson--Wright inequality to establish an optimal restricted isometry property in the presence of coupled terms. Empirical evaluations on synthetic data demonstrate that our algorithm achieves competitive performance relative to state-of-the-art methods. Moreover, we propose a novel notion of matrix incoherence tailored to the EDG setting and provide robustness guarantees for our method.
Similar Papers
Provable Non-Convex Euclidean Distance Matrix Completion: Geometry, Reconstruction, and Robustness
Optimization and Control
Finds object shapes from partial distance clues.
Decentralized Online Riemannian Optimization Beyond Hadamard Manifolds
Optimization and Control
Makes smart machines learn better on curved paths.
Guaranteed Noisy CP Tensor Recovery via Riemannian Optimization on the Segre Manifold
Machine Learning (Stat)
Finds hidden patterns in messy data.