CLeaRForecast: Contrastive Learning of High-Purity Representations for Time Series Forecasting
This work addresses noise issues in time series forecasting for domains like finance or weather, but it appears incremental as it builds on existing contrastive learning paradigms with specific improvements.
The paper tackles the problem of noise in time series forecasting by proposing CLeaRForecast, a contrastive learning framework that learns high-purity representations through sample, feature, and architecture purifying methods, resulting in superior performance in various downstream tasks.
Time series forecasting (TSF) holds significant importance in modern society, spanning numerous domains. Previous representation learning-based TSF algorithms typically embrace a contrastive learning paradigm featuring segregated trend-periodicity representations. Yet, these methodologies disregard the inherent high-impact noise embedded within time series data, resulting in representation inaccuracies and seriously demoting the forecasting performance. To address this issue, we propose CLeaRForecast, a novel contrastive learning framework to learn high-purity time series representations with proposed sample, feature, and architecture purifying methods. More specifically, to avoid more noise adding caused by the transformations of original samples (series), transformations are respectively applied for trendy and periodic parts to provide better positive samples with obviously less noise. Moreover, we introduce a channel independent training manner to mitigate noise originating from unrelated variables in the multivariate series. By employing a streamlined deep-learning backbone and a comprehensive global contrastive loss function, we prevent noise introduction due to redundant or uneven learning of periodicity and trend. Experimental results show the superior performance of CLeaRForecast in various downstream TSF tasks.