时间序列预测在众多领域中(例如电力、能源、天气、交通等)都有广泛的应用。时间序列预测问题极具挑战性,尤其是长程时间序列预测(long-term series forecasting)。在长程时间序列预测中,需要根据现有的数据对未来做出较长时段的预测。在部分场景中,模型输出的长度可以达到 1000 以上,覆盖若干周期。该问题对预测模型的精度和计算效率均有较高的要求。且时间序列往往会受到分布偏移和噪音的影响,使得预测难度大大增加。
[1] [Time-series Transformer Survey] Qingsong Wen, Tian Zhou, Chaoli Zhang, Weiqi Chen, Ziqing Ma, Junchi Yan, Liang Sun, “Transformers in Time Series: A Survey,” arXiv preprint arXiv:2202.07125 (2022). Website: https://github.com/qingsongedu/time-series-transformers-review
[2] [KDD’22 Quatformer] Weiqi Chen, Wenwei Wang, Bingqing Peng, Qingsong Wen, Tian Zhou, Liang Sun, “Learning to Rotate: Quaternion Transformer for Complicated Periodical Time Series Forecasting”, in Proc. 28th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD’22), Washington DC, Aug. 2022.
[3] [KDD’22 Tutorial] Qingsong Wen, Linxiao Yang, Tian Zhou, Liang Sun, “Robust Time Series Analysis and Applications: An Industrial Perspective,” in the 28th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD’22), Washington DC, USA, Aug. 14-18, 2022. Website: https://qingsongedu.github.io/timeseries-tutorial-kdd-2022/
[4] [IJCAI’22 Tutorial] Qingsong Wen, Linxiao Yang, Tian Zhou, Liang Sun, “Robust Time Series Analysis: from Theory to Applications in the AI Era,” in the 31st International Joint Conference on Artificial Intelligence (IJCAI 2022), Vienna, Austria, Jul. 23-29, 2022. Website: https://sites.google.com/view/timeseries-tutorial-ijcai-2022
[5] [招聘全职/实习生] 阿里达摩院DI Lab - 常年招全职/实习生: AI for Time Series, AIOps, XAI等方向 (杭州/西雅图) JD: https://zhuanlan.zhihu.com/p/528948916