Paper reading: a perspective view and survey of meta-learning
Goal: build a self-adaptive learning algorithm.
Idea: improve the learning bias dynamically through experience by the continuous accumulation of meta-knowledge
雖然不同學者對于meta learning定義不同,但是有一點是相同的:How can we exploit knowledge about learning (i.e., meta-knowledge) to improve the performance of learning algorithms.
Meta learning advocates the need for continuous adaptation of the learner at different levels of abstraction.