Figure 2From: Inferring multi-target QSAR models with taxonomy-based multi-task learningKnowledge transfer in multi-task learning. Illustration of a knowledge transfer from task 2, which comprises a lot of training data (green), to a similar task 1, which contains little training data (blue). The ∊-insensitive tubes around the regression functions w1Tx and w2Tx are colored gray. The regularizer J( w 1 ,…, w T ) forces the model of task 1 (w1) to be more similar to the model of task 2 (w2). A model w1 that is more similar to w2 predicts the unknown data (red) better, which results in a better generalization of the model.Back to article page