Inferential Theory of Learning
The Inferential Theory of Learning (ITL) is a means of classifying and understanding learning processes, both cognitive and machine, by the types of inference they make and by the way knowledge is created and transformed through learning. Such a view of learning processes is in contrast to the Computational Theory of Learning, in which learning strategies are organized by their computational complexity. Through the Inferential Theory, a learning process can be described as one or more knowledge transmutations (e.g., induction, abstraction, similization).
The Inferential Theory of Learning (ITL) was proposed by Michalski (1991, 1994) as a unified framework for developing and implementing multistrategy learning systems. ITL recognizes a learning process as consisting of three components: the input facts, the background knowledge, and the inference strategy being employed to generate new knowledge and thereby enhance...
- Michalski, R. S. (1991). Inferential learning theory as a basis for multistrategy task-adaptive learning. Proceedings of the First International Workshop on Multistrategy Learning (pp. 3–18). Harpers Ferry, WV.Google Scholar
- Michalski, R. S. (1994). Inferential learning theory: developing foundations for multistrategy learning. In R. S. Michalski & G. Tecuci (Eds.), Machine learning – A multistrategy approach (pp. 3–61). San Mateo: Morgan Kaufmann.Google Scholar