Learning Stable Linear Dynamical Systems with the Weighted Least Square Method

Authors: Wenbing Huang, Lele Cao, Fuchun Sun, Deli Zhao, Huaping Liu, Shanshan Yu

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

Reproducibility Variable Result LLM Response
Research Type Experimental Comparative experimental evaluations demonstrate that our proposed methods outperform the state-of-the-art methods regarding the reconstruction accuracy and the learning efficiency.
Researcher Affiliation Academia Wenbing Huang1, Lele Cao1, Fuchun Sun1, Deli Zhao, Huaping Liu1 and Shanshan Yu 1 Department of Computer Science and Technology, Tsinghua University, State Key Lab. of Intelligent Technology and Systems, Tsinghua National Lab. for Information Science and Technology (TNList);
Pseudocode Yes Algorithm 1: Learning the stable transition matrix by WLS
Open Source Code No The paper does not provide concrete access to source code for the methodology described in this paper.
Open Datasets Yes The UCLA dataset [Saisan et al., 2001] contains 50 categories of dynamic textures... The traffic dataset UCSD [Chan and Vasconcelos, 2005] consists of 254 video sequences of highway traffic... Cambridge [Kim and Cipolla, 2009] consists of 900 images sequences of 9 gesture classes...
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning for training, validation, or testing.
Hardware Specification Yes All experiments are carried out with Matlab 8.1.0.604 (R2013a) on Intel Core i5, 2.20-GHz CPU with 12-GB RAM.
Software Dependencies Yes All experiments are carried out with Matlab 8.1.0.604 (R2013a) on Intel Core i5, 2.20-GHz CPU with 12-GB RAM. As a convex problem, QP can be solved efficiently with the M function embedded in the Matlab software, i.e. quadprog( ).
Experiment Setup Yes In this experiment, the hidden dimension of the LDS model n is set to be 40. As for WLS and DWLS, the parameter threshold in Algorithm 1 is fixed to be 0.0001.