On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

9 results for “subspace-constrained”

Linear Convergence of the Subspace Constrained Mean Shift Algorithm: From Euclidean to Directional Datastat.ML

Paper proves linear convergence of SCMS algorithm for directional data.

problem Identifying density ridges in directional data.

method Generalized SCMS algorithm to directional data, derived from SCGA with adaptive step size.

result Linear convergence of the proposed directional SCMS algorithm.

mean shift algorithm density ridge directional data linear convergence subspace constrained

2021-04-29abs ↗pdf ↗

Theoretical Guarantees for the Subspace-Constrained Tyler's Estimatormath.ST

Theoretical guarantees for STE, a robust subspace recovery method.

problem Recovering a low-dimensional subspace from corrupted data.

method Subspace-constrained Tyler's estimator (STE) with initialization conditions.

result STE can effectively recover the subspace under certain conditions.

robust subspace recovery Tyler's estimator subspace-constrained outliers inlier-outlier model

2024-03-27abs ↗pdf ↗

Nonconvex Matrix Completion with Linearly Parameterized Factorsmath.ST

Unified framework for nonconvex matrix completion with linearly parameterized factors.

problem Matrix completion with improved accuracy using linearly parameterized factors.

method Unified nonconvex optimization framework with Correlated Parametric Factorization condition.

result Uniform upper bounds for low-rank estimation at any local minimum.

matrix completion nonconvex optimization linearly parameterized factors Correlated Parametric Factorization low-rank estimation

2020-03-29abs ↗pdf ↗

Optimal Structured Principal Subspace Estimation: Metric Entropy and Minimax Ratesmath.ST

Unified framework for structured principal subspace estimation with bounds and rates.

problem Structured principal subspace estimation problems.

method Unified framework, minimax lower and upper bounds, information-geometric complexity.

result Minimax rates of convergence for specific settings, including optimal rates for non-negative PCA/SVD.

principal subspace structured estimation minimax bounds non-negative PCA sparse PCA

2020-02-18abs ↗pdf ↗

Algorithms for ridge estimation with convergence guaranteesstat.ML

Two new algorithms recover ridge lines from point clouds with convergence guarantees.

problem Extracting filamentary structure from point clouds.

method Proposes two novel algorithms with convergence guarantees.

result The algorithms can asymptotically recover the full ridge set.

ridge estimation convergence guarantees point cloud filamentary structure density ridges

2021-04-26abs ↗pdf ↗

Mode and Ridge Estimation in Euclidean and Directional Product Spaces: A Mean Shift Approachstat.ML

Estimates modes and ridges in mixed Euclidean and directional spaces.

problem Estimating local modes and density ridges in product spaces combining Euclidean and directional metrics.

method Extends mean shift algorithm to product spaces, addressing challenges in generalization.

result Established convergence of the proposed methods and demonstrated effectiveness on real-world datasets.

mean shift product spaces local modes density ridges Euclidean

2021-10-16abs ↗pdf ↗

Exact Subspace Diffusion for Decentralized Multitask Learningcs.LG

New algorithm improves multitask learning across diverse agents.

problem Performance degradation in decentralized learning with heterogeneous objectives.

method Developed an exact subspace diffusion algorithm for multitask learning over networks.

result The algorithm outperforms alternatives in noisy gradient approximations.

multitask learning decentralized learning subspace diffusion consensus mechanisms performance degradation

2023-04-14abs ↗pdf ↗

Space Partitioning and Regression Mode Seeking via a Mean-Shift-Inspired Algorithmstat.ML

Develops an MS-inspired algorithm for regression mode finding and space partitioning.

problem Finding local modes of regression functions and partitioning input space.

method Mean-shift-inspired algorithm for iterative gradient ascent.

result Proves convergence and rates of convergence for estimated local modes.

mean shift regression mode space partitioning gradient ascent nonparametric methods

2021-04-20abs ↗pdf ↗

Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learningcs.LG

New method prevents forgetting in LLMs by dynamically identifying task-specific subspaces.

problem Catastrophic forgetting in continual learning of LLMs.

method Adaptive Singular Value Decomposition (SVD) for constrained full fine-tuning.

result Achieves state-of-the-art results in continual learning benchmarks.

continual learning large language models catastrophic forgetting adaptive SVD dynamic subspaces

2025-04-09abs ↗pdf ↗