Synthetic Time Series

Synthetic time series generation focuses on creating artificial datasets that mimic real-world temporal patterns, primarily to address data scarcity, enhance model training, and improve the interpretability of time series analysis. Current research emphasizes the development and evaluation of generative models, including GANs, diffusion models, and transformer-based architectures, often incorporating techniques like wavelet transforms and conditional generation to control the characteristics of the synthetic data. This field is crucial for advancing machine learning applications in various domains, from anomaly detection in cloud computing and forecasting in finance to improving the accuracy and efficiency of time series classification in neuroscience and other data-limited areas.

Papers

September 22, 2022

StyleTime: Style Transfer for Synthetic Time Series Generation
Yousef El-Laham, Svitlana Vyetrenko
Synthetic Data Style Transfer Image Stylization Neural Style Transfer Synthetic Time Series

July 16, 2022

Transfer learning for time series classification using synthetic data generation
Yarden Rotem, Nathaniel Shimoni, Lior Rokach, Bracha Shapira
Synthetic Data Generation Formality Transfer Time Series Classification Classification Task Regression Task Synthetic Time Series

June 28, 2022

TTS-CGAN: A Transformer Time-Series Conditional GAN for Biosignal Data Augmentation
Xiaomin Li, Anne Hee Hiong Ngu, Vangelis Metsis
Generative Adversarial Network Data Augmentation Synthetic Data Transformer Based State of the Art GAN Conditional GAN Synthetic Time Series

June 18, 2022

Scalable Classifier-Agnostic Channel Selection for Multivariate Time Series Classification
Bhaskar Dhariyal, Thach Le Nguyen, Georgiana Ifrim
Time Series Classification Multivariate Time Series Classification Synthetic Time Series Channel Selection

June 2, 2022

Generating Sparse Counterfactual Explanations For Multivariate Time Series
Jana Lang, Martin Giese, Winfried Ilg, Sebastian Otte
Generative Adversarial Network Counterfactual Explanation Multivariate Time Series Counterfactual Prediction Synthetic Time Series Sparse Counterfactual Explanation

May 23, 2022

Time-series Transformer Generative Adversarial Networks
Padmanaba Srinivasan, William J. Knottenbelt
Synthetic Data Time Series Autoregressive Model Synthetic Time Series

April 13, 2022

A quantum generative model for multi-dimensional time series using Hamiltonian learning
Haim Horowitz, Pooja Rao, Santosh Kumar Radha
Time Series Quantum Machine Learning Quantum Physic Quantum Computer Hamiltonian Learning Synthetic Time Series Quantum Dynamic Quantum Generative Adversarial Network

March 8, 2022

Sparsification and Filtering for Spatial-temporal GNN in Multivariate Time-series
Yuanrong Wang, Tomaso Aste
Spatio Temporal Multivariate Time Series Time Series Prediction Random Sparsification Spatial Temporal Graph Online Filtering Synthetic Time Series Correlation Graph

January 13, 2022

Privacy Amplification by Subsampling in Time Domain
Tatsuki Koga, Casey Meehan, Kamalika Chaudhuri
Differential Privacy Time Series Data Privacy Amplification Synthetic Time Series Additive Noise Privacy Mechanism

January 12, 2022

Data augmentation through multivariate scenario forecasting in Data Centers using Generative Adversarial Networks
Jaime Pérez, Patricia Arroba, José M. Moya
Generative Adversarial Network Data Augmentation Synthetic Data Data Center Synthetic Time Series Based Augmentation

Synthetic Time Series

Papers

StyleTime: Style Transfer for Synthetic Time Series Generation

Transfer learning for time series classification using synthetic data generation

TTS-CGAN: A Transformer Time-Series Conditional GAN for Biosignal Data Augmentation

Scalable Classifier-Agnostic Channel Selection for Multivariate Time Series Classification

Generating Sparse Counterfactual Explanations For Multivariate Time Series

Time-series Transformer Generative Adversarial Networks

A quantum generative model for multi-dimensional time series using Hamiltonian learning

Sparsification and Filtering for Spatial-temporal GNN in Multivariate Time-series

Privacy Amplification by Subsampling in Time Domain

Data augmentation through multivariate scenario forecasting in Data Centers using Generative Adversarial Networks