A Perspective View And Survey Of Meta-Learning

The first part of this paper provides a perspective view of meta-learning in which the goal is to build self-adaptive learning algorithms. The idea is to improve the learning bias dynamically through experience by the continuous accumulation of meta-knowledge. The second part of this paper provides a survey of meta-learning as reported by the machine-learning literature. We find that different researchers hold different views of what the term meta-learning exactly means. Despite different views and research lines, a question remains constant: how can we exploit knowledge about learning (i.e., meta-knowledge) to improve the performance of learning algorithms? Clearly the answer to this question is key to the advancement of the field and continues being the subject of intensive research.