Abstract:Time series is a typical data type in numerous domains; however, labeling large amounts of time series data can be costly and time-consuming. Learning effective representation from unlabeled time series data is a challenging task. Contrastive learning stands out as a promising method to acquire representations of unlabeled time series data. Therefore, we propose a self-supervised time-series representation learning framework via Time-Frequency Fusion Contrasting (TF-FC) to learn time-series representation from unlabeled data. Specifically, TF-FC combines time-domain augmentation with frequency-domain augmentation to generate the diverse samples. For time-domain augmentation, the raw time series data pass through the time-domain augmentation bank (such as jitter, scaling, permutation, and masking) and get time-domain augmentation data. For frequency-domain augmentation, first, the raw time series undergoes conversion into frequency domain data following Fast Fourier Transform (FFT) analysis. Then, the frequency data passes through the frequency-domain augmentation bank (such as low pass filter, remove frequency, add frequency, and phase shift) and gets frequency-domain augmentation data. The fusion method of time-domain augmentation data and frequency-domain augmentation data is kernel PCA, which is useful for extracting nonlinear features in high-dimensional spaces. By capturing both the time and frequency domains of the time series, the proposed approach is able to extract more informative features from the data, enhancing the model's capacity to distinguish between different time series. To verify the effectiveness of the TF-FC method, we conducted experiments on four time series domain datasets (i.e., SleepEEG, HAR, Gesture, and Epilepsy). Experimental results show that TF-FC significantly improves in recognition accuracy compared with other SOTA methods.

Towards Unsupervised Time Series Representation Learning: A Decomposition Perspective

Multi-view Self-Supervised Contrastive Learning for Multivariate Time Series

Time-Series Representation Learning via Temporal and Contextual Contrasting

DCS: Debiased Contrastive Learning with Weak Supervision for Time Series Classification

Self-Supervised Contrastive Representation Learning for Semi-Supervised Time-Series Classification

Unsupervised Representation Learning for Time Series: A Review

Simple Contrastive Representation Learning for Time Series Forecasting

Multi-Task Self-Supervised Time-Series Representation Learning

SimTS: Rethinking Contrastive Representation Learning for Time Series Forecasting

Time Series Representation Learning with Supervised Contrastive Temporal Transformer

Unsupervised Time-Series Representation Learning with Iterative Bilinear Temporal-Spectral Fusion.

Semi-supervised contrastive learning with decomposition-based data augmentation for time series classification

TS-TWC: A time series representation learning framework based on Time-Wavelet contrasting

TS2Vec: Towards Universal Representation of Time Series

Time-series representation learning via Time-Frequency Fusion Contrasting

CLeaRForecast: Contrastive Learning of High-Purity Representations for Time Series Forecasting

Enabling Tensor Decomposition for Time-Series Classification via A Simple Pseudo-Laplacian Contrast

CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting

Distillation Enhanced Time Series Forecasting Network with Momentum Contrastive Learning