Tensor Decompositions for Learning Latent Variable Models

Authors: Animashree Anandkumar, Rong Ge, Daniel Hsu, Sham M. Kakade, Matus Telgarsky

JMLR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We provide a convergence analysis of this method for orthogonally decomposable symmetric tensors, as well as a detailed perturbation analysis for a robust (and a computationally tractable) variant (Theorem 5.1). This perturbation analysis can be viewed as an analogue of Wedin s perturbation theorem for singular vectors of matrices (Wedin, 1972)
Researcher Affiliation Collaboration Animashree Anandkumar EMAIL Electrical Engineering and Computer Science University of California, Irvine... Rong Ge EMAIL Microsoft Research...
Pseudocode Yes Algorithm 1 Robust tensor power method input symmetric tensor T Rk k k, number of iterations L, N. output the estimated eigenvector/eigenvalue pair; the deflated tensor.
Open Source Code No The paper does not explicitly state that code is made available, nor does it provide any links to source code repositories.
Open Datasets No The paper discusses various latent variable models and their theoretical properties but does not describe experiments performed on specific datasets or provide access information for any datasets.
Dataset Splits No The paper focuses on theoretical analysis and algorithm design for latent variable models, and thus does not include details on dataset splits for experimental reproduction.
Hardware Specification No The paper focuses on theoretical computational complexity and algorithm analysis, and therefore does not provide any specific hardware details used for running experiments.
Software Dependencies No The paper is theoretical and does not describe implementation details or specific software dependencies with version numbers used in the authors' work.
Experiment Setup No The paper presents theoretical methods and analyses without conducting empirical experiments, and thus does not include details on experimental setup or hyperparameter configurations.