Efficiently escaping saddle points on manifolds

Authors: Christopher Criscitiello, Nicolas Boumal

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Generalizing Jin et al. s recent work on perturbed gradient descent (PGD) for optimization on linear spaces [...], we propose a version of perturbed Riemannian gradient descent (PRGD) to show that necessary optimality conditions can be met approximately with high probability, without evaluating the Hessian. ... This matches the complexity of PGD in the Euclidean case. Crucially, the dependence on dimension is low. ... The algorithm requires knowledge of the Lipschitz constants defined below, which makes this a mostly theoretical algorithm but see Appendix D for explicit constants in the case of PCA.
Researcher Affiliation Academia Chris Criscitiello Department of Mathematics Princeton University Princeton, NJ 08544 ccriscitiello6@gmail.com Nicolas Boumal Department of Mathematics Princeton University Princeton, NJ 08544 nboumal@math.princeton.edu
Pseudocode Yes Algorithm 1 PRGD(x0, , r, T , , T, b) ... procedure TANGENTSPACESTEPS(x, s0, , b, T )
Open Source Code No No explicit statement or link providing access to the source code for the described methodology was found. The paper indicates its theoretical nature: 'The algorithm requires knowledge of the Lipschitz constants defined below, which makes this a mostly theoretical algorithm but see Appendix D for explicit constants in the case of PCA.'
Open Datasets No The paper is theoretical and does not conduct experiments, therefore no dataset information is provided. It mentions applications like 'PCA and low-rank matrix completion' but these are not datasets used in empirical evaluation.
Dataset Splits No The paper is theoretical and does not conduct experiments with data, therefore no information about dataset splits for training, validation, or testing is provided.
Hardware Specification No The paper is theoretical and does not describe empirical experiments, therefore no hardware specifications are provided.
Software Dependencies No The paper is theoretical and does not describe empirical experiments or specific implementations, therefore no software dependencies with version numbers are provided.
Experiment Setup No The paper is theoretical and does not describe empirical experiments, therefore no experimental setup details such as hyperparameters or training configurations are provided.