Meta-Learning Enhanced Trade Forecasting: A Neural Framework Leveraging Efficient Multicommodity STL Decomposition

Figure 1

The architecture of the proposed Meta-TFSTL. FC: fully connected layer. In ANIL, the model’s feature-extracting backbone remains static (Regular Optimization), while the output head undergoes gradient descent updates (Meta-Learning-Based Optimization). This ensures stable foundational representations while allowing rapid task-specific adaptations.