Generalized Singular Value Thresholding

Authors: Canyi Lu, Changbo Zhu, Chunyan Xu, Shuicheng Yan, Zhouchen Lin

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments In this section, we conduct some experiments on the matrix completion problem to test our proposed GPG algorithm... We compared GPG with IRNN on both synthetic and real data.
Researcher Affiliation Academia Canyi Lu1, Changbo Zhu1, Chunyan Xu2, Shuicheng Yan1, Zhouchen Lin3, 1 Department of Electrical and Computer Engineering, National University of Singapore 2 School of Computer Science and Technology, Huazhong University of Science and Technology 3 Key Laboratory of Machine Perception (MOE), School of EECS, Peking University
Pseudocode Yes Algorithm 1: A General Solver to (5) in which g satisfying Assumption 1
Open Source Code No The paper does not provide any specific statements about releasing source code or links to a code repository for the described methodology.
Open Datasets Yes We test on the Movie Lens data set (Herlocker et al. 1999) which includes three problems, movie-100K , movie-1M and movie-10M .
Dataset Splits No The paper describes data generation and missing data scenarios (e.g., 'Half of the elements in M are missing', '40% of pixels are uniformly missing') but does not specify exact training, validation, or test dataset splits with percentages or sample counts.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments.
Software Dependencies No The paper refers to algorithms and methods (e.g., ALM, APGL) but does not specify any software dependencies with version numbers.
Experiment Setup Yes The initial value of λ in the Logarithm penalty is set to λ0, and dynamically decreased till reaching λt. We set λ0 = 0.9||PΩ(M)|| , and λt = 10 5λ0. For this task, we set λ0 = 10||PΩ(M)|| , and λt = 0.1λ0 in GPG.