Learning from Multiway Data: Simple and Efficient Tensor Regression

Authors: Rose Yu, Yan Liu

ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate superior empirical performance on both multi-linear multi-task learning and spatio-temporal applications.
Researcher Affiliation Academia Rose Yu QIYU@USC.EDU Yan Liu YANLIU.CS@USC.EDU Department of Computer Science, University of Southern California
Pseudocode Yes Algorithm 1 Subsampled Tensor Projected Gradient; Algorithm 2 Iterative Tensor Projection (ITP)
Open Source Code No The paper does not provide any statements or links indicating that its source code is available.
Open Datasets Yes The U.S. Historical Climatology Network (USHCN) daily (http://cdiac.ornl. gov/ftp/ushcn_daily/) contains daily measurements for 5 climate variables for more than 100 years.
Dataset Splits Yes We also select 200 instances as the validation set.
Hardware Specification No The paper does not provide specific details about the hardware used for running the experiments (e.g., GPU/CPU models, memory specifications).
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., Python version, specific library versions).
Experiment Setup Yes We choose VAR (3) model and use 5-fold cross-validation to select the rank during the training phase. For both datasets, we normalize each individual time series by removing the mean and dividing by standard deviation. The model parameters are selected by minimizing the mean squared error on the validation set.