Fig. 6: Visualize the weights from the adaptive task module across three benchmark datasets (BindingDB, KIBA, and Davis). | Nature Communications

Fig. 6: Visualize the weights from the adaptive task module across three benchmark datasets (BindingDB, KIBA, and Davis).

From: A meta learning and task adaptive approach for drug target affinity prediction

Fig. 6: Visualize the weights from the adaptive task module across three benchmark datasets (BindingDB, KIBA, and Davis).The alternative text for this image may have been generated using AI.

The horizontal axis represents the gradient similarity between the support and query sets of each task, and the vertical axis represents the query loss. Each scatter point corresponds to a specific task, and the color of the point indicates its assigned sampling weight: warmer colors (red/orange) denote higher weights, while cooler colors (blue/cyan) denote lower weights. Source data are provided as a Source Data file.

Back to article page