Generalized Higher-Order Tensor Decomposition via Parallel ADMM

Authors: Fanhua Shang, Yuanyuan Liu, James Cheng

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results verify that our regularized formulation is effective, and our methods are robust to noise or outliers. In this section, we evaluate both the effectiveness and efficiency of our methods for solving tensor decomposition problems using both synthetic and real-world data. All experiments were performed on an Intel(R) Core (TM) i5-4570 (3.20 GHz) PC running Windows 7 with 8GB main memory.
Researcher Affiliation Academia 1Department of Computer Science and Engineering, The Chinese University of Hong Kong 2Department of Systems Engineering and Engineering Management, The Chinese University of Hong Kong
Pseudocode Yes Algorithm 1 Solving problem (5) via parallel ADMM; Algorithm 2 Solving problem (11) via Parallel ADMM
Open Source Code No The paper does not provide any explicit statement about releasing its source code or a link to a code repository.
Open Datasets Yes MRI Data This part compares our CTD and NCTD methods, HOSVD and HOOI on a 181 217 181 brain MRI data used in (Liu et al., 2009).
Dataset Splits No The paper describes how synthetic data was generated and mentions a real-world MRI dataset, but it does not specify explicit training/validation/test splits or percentages.
Hardware Specification Yes All experiments were performed on an Intel(R) Core (TM) i5-4570 (3.20 GHz) PC running Windows 7 with 8GB main memory.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, or frameworks with their versions).
Experiment Setup Yes We set the regularization parameter λ = 100 for our methods. Other parameters of Mixture, ADMM, HOSVD and HOOI are set to their default values. We set the tensor ranks Rn = 1.2r , n = 1, . . . , N and Tol = 10 5 for all these algorithms. Initialize: Y 0 n = 0, G0 n = 0, U 0 n = rand(In, Rn), µ0 = 10 4, µmax = 1010, ρ = 1.05, and Tol = 10 5.