Spectral-Aware Reservoir Computing for Fast and Accurate Time Series Classification

Analyzing inherent temporal dynamics is a critical pathway for time series classification, where Reservoir Computing (RC) exhibits effectiveness and high efficiency. However, typical RC considers recursive updates from adjacent states, struggling with long-term dependencies. In response, this paper proposes a Spectral-Aware Reservoir Computing framework (SARC), incorporating spectral insights to enhance long-term dependency modeling. Prominent frequencies are initially extracted to reveal explicit or implicit cyclical patterns. For each prominent frequency, SARC further integrates a Frequency-informed Reservoir Network (FreqRes) to adequately capture both sequential and cyclical dynamics, thereby deriving effective dynamic features. Synthesizing these features across various frequencies, SARC offers a multi-scale analysis of temporal dynamics and improves the modeling of long-term dependencies. Experiments on public datasets demonstrate that SARC achieves state-of-the-art results, while maintaining high efficiency compared to existing methods.

Time series data, which tracks changes over time like heartbeats or stock prices, often contains hidden cyclical patterns crucial for accurate classification. However, existing methods struggle to capture these patterns effectively. Our work introduces SARC, a new approach that identifies cyclical patterns across multiple time spans, like daily or seasonal cycles. Using a static model, SARC eliminates the need for heavy computational resources. Experiments on real-world datasets show that SARC achieves superior classification accuracy and higher efficiency than multiple existing methods. This makes it a practical tool for applications requiring quick, reliable decisions, such as detecting medical conditions by categorizing physiological signals.