Reproduce “TimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series Analytics” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Peng Chen, Yihang Wang, Zhongwen Rao, Yang Shu, Yuying Qiu, Yuxuan Chen, Lujia Pan, Bin Yang and 1 more
TimeSeek is a pretrained time-series model for forecasting, anomaly detection, and classification that selectively shares information across temporal scales and channels. Its shared feature extractor adapts scale fusion and channel modeling, while task-specific components produce each task’s outputs. In zero-shot Weather forecasting at horizon 96, it reports lower MSE and faster inference than Time-MOE, but is not fastest or smallest.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.