Graphical Time Warping for Joint Alignment of Multiple Curves

Authors: Yizhi Wang, David J. Miller, Kira Poskanzer, Yue Wang, Lin Tian, Guoqiang Yu

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

Reproducibility Variable Result LLM Response
Research Type Experimental We used synthetic and real data to compare the performance of GTW and DTW. For the synthetic data, we evaluate the performance by the estimation error for the warping path Pn. For real data, we examine the spatial delay pattern relative to a reference curve.
Researcher Affiliation Academia Yizhi Wang Virginia Tech yzwang@vt.edu David J. Miller Pennsylvania State University djmiller@engr.psu.edu Kira Poskanzer University of California, San Francisco Kira.Poskanzer@ucsf.edu Yue Wang Virginia Tech yuewang@vt.edu Lin Tian University of California, Davis lintian@ucdavis.edu Guoqiang Yu Virginia Tech yug@vt.edu
Pseudocode No The paper includes a flowchart (Fig.1d) describing the GTW algorithm steps, but does not provide structured pseudocode or an explicitly labeled algorithm block.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the methodology is openly available.
Open Datasets Yes We applied GTW to estimate the propagation patterns of astrocyte calcium fluorescent imaging data [22, 8]. The movie was obtained from a neuro-astrocyte co-cultured Down syndrome cell line.
Dataset Splits No The paper mentions hyperparameters can be tuned via 'cross validation' but does not provide specific details on training, validation, or test splits for the experiments conducted.
Hardware Specification No The paper does not provide any specific hardware details (e.g., CPU/GPU models, memory specifications) used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, or specific solver versions).
Experiment Setup Yes All hyperparameters were initialized to 0; the noise variance was initialized to 0.01. In addition, the distance of the path from the diagonal line was penalized via a hyperparameter β = d/σ2, where d is the distance of a point in the path to the diagonal. When the parameter and hyperparameter changes were all less than 0.001, we stopped the algorithm. Convergence usually occurred within 10 iterations.