AdaRNN: Adaptive Learning and Forecasting of Time Series

By Yuntao Du, Jindong Wang, Wenjie Feng, Sinno Jialin Pan, Tao Qin, Renjun Xu, Chongjun Wang. International Conference on Information and Knowledge Management, 2021

371 citations recorded by the source corpus.

Time series has wide applications in the real world and is known to be difficult to forecast. Since its statistical properties change over time, its distribution also changes temporally, which will cause severe distribution shift problem to existing methods. However, it remains unexplored to model the time series in the distribution perspective. In this paper, we term this as Temporal Covariate Shift (TCS). This paper proposes Adaptive RNNs (AdaRNN) to tackle the TCS problem by building an adaptive model that generalizes well on the unseen test data. AdaRNN is sequentially composed of two novel algorithms. First, we propose Temporal Distribution Characterization to better characterize the distribution information in the TS. Second, we propose Temporal Distribution Matching to reduce the distribution mismatch in TS to learn the adaptive TS model. AdaRNN is a general framework with flexible distribution distances integrated. Experiments on human activity recognition, air quality prediction, and financial analysis show that AdaRNN outperforms the latest methods by a classification accuracy of 2.6% and significantly reduces the RMSE by 9.0%. We also show that the temporal distribution matching algorithm can be extended in Transformer structure to boost its performance.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/6ef770d11e3e5918646b2f5a97c0bcc8bc9b9256

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Junbo Zhang
2. Yu Zheng
3. Tao Qin
4. Jindong Wang
5. J. Choo
6. Yidong Ouyang
7. Zezhi Shao
8. Chang Liu

In Signal Processing

1. Junbo Zhang
2. Tao Qin
3. Jindong Wang
4. J. Choo
5. Zezhi Shao
6. Yu Zheng
7. Cuiling Lan
8. Xing Xie

Citations in the corpus

• Reversible Instance Normalization for Accurate Time-Series Forecasting against Distribution Shift (2022, background): “Domain adaptation (Tzeng et al., 2017; Ganin et al., 2016; Wang et al., 2018) and domain generalization (Wang et al., 2021; Li et al., 2018; Muandet et al., 2013) are common ways to alleviate the distribution shift.”
• Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Survey (2023, background): “While several studies [227] investigated defeating distribution shifts in time series, they fail to encode the spatial correlations among locations.”
• Generalizing to Unseen Domains: A Survey on Domain Generalization (2021, methodology): “…learning: DANN [91, 92]/ MMD-AAE [72]/ DLOW [93]/ CIAN [94]/ ER [95]/ MADDG [96]/ FSUS [45]/ MMLD [97]/ DANNCE [98]/ CAADG [99]/ SSDG [100]/ AdaRNN [80] ASR-Norm [81]/ SADG [101]/ [102, 103] Invariant risk minimization: IRM [104]/IB-IRM [105]/VRex [106]/ReLIC [107]/[108, 109, 110, 111, 112]”