Cross-Domain Joint-Embedding Pretraining For Heterogeneous Industrial Time Series
The authors propose JEPA-HITS, a channel-independent joint-embedding predictive pretraining method for heterogeneous industrial time series. They learn one shared JEPA encoder across battery, CMP, CNC, and bearing data while preserving dataset-specific temporal scales, using leave-one-domain-out (LODO) pretraining to test transfer to unseen domains. Compared to the same architecture with random initialization, LODO pretraining reduces RMSE on NASA/CMP/CNC and improves bearing fault macro-F1.
