Voice2Series: Reprogramming Acoustic Models for Time Series Classification

17 Jun 2021  ·  Chao-Han Huck Yang, Yun-Yun Tsai, Pin-Yu Chen ·

Learning to classify time series with limited data is a practical yet challenging problem. Current methods are primarily based on hand-designed feature extraction rules or domain-specific data augmentation. Motivated by the advances in deep speech processing models and the fact that voice data are univariate temporal signals, in this paper, we propose Voice2Series (V2S), a novel end-to-end approach that reprograms acoustic models for time series classification, through input transformation learning and output label mapping. Leveraging the representation learning power of a large-scale pre-trained speech processing model, on 30 different time series tasks we show that V2S performs competitive results on 19 time series classification tasks. We further provide a theoretical justification of V2S by proving its population risk is upper bounded by the source risk and a Wasserstein distance accounting for feature alignment via reprogramming. Our results offer new and effective means to time series classification.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Time Series Classification Earthquakes V2Sa Accuracy (Test) 78.42 # 1
Time Series Classification FordA V2Sa Acc. (test) 100 # 1
ECG Classification UCR Time Series Classification Archive V2Sa Accuracy (Test) 93.96 # 1

Methods


No methods listed for this paper. Add relevant methods here