Time-series Generative Adversarial Networks

Authors: Jinsung Yoon, Daniel Jarrett, Mihaela van der Schaar

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, we evaluate the ability of our method to generate realistic samples using a variety of real and synthetic time-series datasets. Qualitatively and quantitatively, we find that the proposed framework consistently and significantly outperforms state-of-the-art benchmarks with respect to measures of similarity and predictive ability.
Researcher Affiliation Academia Jinsung Yoon University of California, Los Angeles, USA jsyoon0823@g.ucla.edu Daniel Jarrett University of Cambridge, UK daniel.jarrett@maths.cam.ac.uk Mihaela van der Schaar University of Cambridge, UK University of California, Los Angeles, USA Alan Turing Institute, UK mv472@cam.ac.uk, mihaela@ee.ucla.edu
Pseudocode Yes Algorithm pseudocode and illustrations with additional detail can be found in the Supplementary Materials.
Open Source Code Yes Implementation of Time GAN can be found at https://bitbucket.org/mvdschaar/ mlforhealthlabpub/src/master/alg/timegan/.
Open Datasets No The paper mentions datasets like "daily historical Google stocks data", "UCI Appliances energy prediction dataset", and a "large private lung cancer pathways dataset." While the Google Stocks and UCI Appliances datasets are publicly known, the paper does not provide concrete access information (e.g., specific links, DOIs, or formal citations with authors/year) for them within the main text.
Dataset Splits No The paper mentions evaluating on a "held-out test set" but does not provide specific percentages, sample counts, or a detailed methodology for how the training, validation, and test splits were performed across the datasets.
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types, or memory) used to run the experiments.
Software Dependencies No The paper does not provide specific version numbers for any software libraries, frameworks, or dependencies used in the experiments.
Experiment Setup Yes In practice, we find that Time GAN is not sensitive to λ and η; for all experiments in Section 5, we set λ = 1 and η = 10.