提案されたTSFoolアプローチは、RNNベースの時系列分類における高度に認識しにくい敵対的な時系列を効率的に作成し、既存の手法を効果、効率、認識性の観点で大幅に上回っています。
CrossTimeNet proposes a novel cross-domain SSL learning framework to enhance time series representation through self-supervised pre-training, achieving superior performance in various domains.
The author introduces pre-trained domain foundation models to address overfitting in Time Series Classification, demonstrating superior performance compared to traditional methods.