Skip to content

Latest commit

 

History

History
64 lines (32 loc) · 4.56 KB

wiki.md

File metadata and controls

64 lines (32 loc) · 4.56 KB
Raw

Time-Series Foundation Models (TSFM)

In this section, we highlight the papers, blogs, and open-source code from IBM's TSFM group.

HuggingFace Model releases

PatchTSMixer: https://huggingface.co/docs/transformers/main/en/model_doc/patchtsmixer

PatchTST: https://huggingface.co/docs/transformers/main/en/model_doc/patchtst

Publications

4 KDD, 1 ICLR, 1 AAAI, 1 ICML, 2 preprints, Total citations: 724

  • TST: Zerveas, G., Jayaraman, S., Patel, D., Bhamidipaty, A., & Eickhoff, C. (2021, August). A transformer-based framework for multivariate time series representation learning. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (pp. 2114-2124). (citations: 504)

  • PatchTST: Nie, Y., Nguyen, N. H., Sinthong, P., & Kalagnanam, J. (2022). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers, ICLR 2023, arXiv preprint arXiv:2211.14730 (citations: 118)

  • PatchTSMixer: Vijay Ekambaram, Arindam Jati, Nam Nguyen, Phanwadee Sinthong, Jayant Kalgnanam. TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting, 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '23), Research Track, August 6-10, 2023, Long Beach, CA, USA https://arxiv.org/abs/2306.09364 (New ones, citations: 8)

  • NPF: Ekambaram, V., Manglik, K., Mukherjee, S., Sajja, S. S. K., Dwivedi, S., & Raykar, V. (2020, August). Attention based multi-modal new product sales time-series forecasting. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 3110-3118). (citations: 45)

  • TLAE: Nguyen, N., & Quanz, B. (2021, May). Temporal latent auto-encoder: A method for probabilistic multivariate time series forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, No. 10, pp. 9117-9125). (citations: 48)

  • HPRO: Arindam  Jati, Vijay Ekambaram, Shaonli Pal, Brian Quanz, Wesley M. Gifford, Pavithra Harsha, Stuart Siegel, Sumanta Mukherjee, Chandra Narayanaswami "Hierarchy-guided Model Selection for Time Series Forecasting, 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '23), Research Track, August 6-10, 2023, Long Beach, CA, USA, KDD 23 https://arxiv.org/abs/2211.15092 (New ones, citations: 1)

  • ConCerNet: Wang Zhang, Tsui-Wei Weng, Subhro Das, Alexandre Megretski, Luca Daniel, Lam M. Nguyen. "ConCerNet: A Contrastive Learning Based Framework for Automated Conservation Law Discovery and Trustworthy Dynamical System Prediction", The 40th International Conference on Machine Learning (ICML 2023)

Workshops/Invited Talks/Tutorials:

  • Lam M. Nguyen, Trang H. Tran, Wang Zhang, Subhro Das, Tsui-Wei Weng. "When Machine Learning meets Dynamical Systems: Theory and Applications", Workshop at The 37th Conference on Artificial Intelligence (AAAI 2023)

Preprints:

  • Trang H. Tran, Lam M. Nguyen, Kyongmin Yeo, Nam Nguyen, Dzung Phan, Roman Vaculin, Jayant Kalagnanam. "An End-to-End Time Series Model for Simultaneous Imputation and Forecast", arXiv preprint 2023

  • Anh Duy Nguyen, Trang H. Tran, Hieu H. Pham, Phi Le Nguyen, Lam M. Nguyen. "Learning Robust and Consistent Time Series Representations: A Dilated Inception-Based Approach", arXiv preprint 2023

Blogs

PatchTST: A Breakthrough in Time Series Forecasting (June 20, 2023) https://towardsdatascience.com/patchtst-a-breakthrough-in-time-series-forecasting-e02d48869ccc

PatchTST — A Step Forward in Time Series Forecasting (June 24, 2023) https://pub.towardsai.net/patchtst-a-step-forward-in-time-series-forecasting-13a8e8f53feb

PatchTST for Time Series Forecasting: Original Results and My Single-Channel Experiments (May 17, 2023) https://medium.com/@lalf_klein/patchtst-for-time-series-forecasting-original-results-and-new-single-channel-experiments-f375699f7b91

Top AI/ML Papers of the Week [12/06 - 18/06] highlighting TSMixer https://www.linkedin.com/pulse/top-aiml-papers-week-1206-1806-bruno-miguel-l-silva/

The Return of the Fallen: Transformers for Forecasting (May 25, 2023) https://towardsdatascience.com/the-return-of-the-fallen-transformers-for-forecasting-24f6fec5bc30

TS Foundation Models - The Battle of Time-series Transformers (June 26, 2023) https://www.linkedin.com/pulse/ts-foundation-models-battle-time-series-vijay-ekambaram/

Other OpenSource Models

https://timeseriesai.github.io/tsai/models.tst.html

https://ts.gluon.ai/stable/api/gluonts/gluonts.torch.model.patch_tst.html

https://nixtla.github.io/neuralforecast/models.patchtst.html

https://github.com/yuqinie98/PatchTST

https://github.com/gzerveas/mvts_transformer