Teaching Randomized Learners
The present paper introduces a new model for teaching randomized learners. Our new model, though based on the classical teaching dimension model, allows to study the influence of various parameters such as the learner’s memory size, its ability to provide or to not provide feedback, and the influence of the order in which examples are presented. Furthermore, within the new model it is possible to investigate new aspects of teaching like teaching from positive data only or teaching with inconsistent teachers.
Furthermore, we provide characterization theorems for teachability from positive data for both ordinary teachers and inconsistent teachers with and without feedback.
KeywordsMarkov Decision Process Concept Class Teaching Process Public Output Positive Data
Unable to display preview. Download preview PDF.
- 1.Angluin, D.: Queries and concept learning. Machine Learning 2(4), 319–342 (1988)Google Scholar
- 3.Angluin, D., Kriķis, M.: Teachers, learners and black boxes. In: Proc. of the 10th Annual Conference on Computational Learning Theory, pp. 285–297. ACM Press, New York (1997)Google Scholar
- 4.Angluin, D., Kriķis, M.: Learning from different teachers. Machine Learning 51(2), 137–163 (2003)Google Scholar
- 6.Bertsekas, D.P.: Dynamic Programming and Optimal Control. Athena ScientificGoogle Scholar
- 13.Jain, S., Lange, S., Nessel, J.: Learning of r.e. languages from good examples. In: Li, M. (ed.) ALT 1997. LNCS (LNAI), vol. 1316, pp. 32–47. Springer, Heidelberg (1997)Google Scholar
- 14.Madani, O., Hanks, S., Condon, A.: On the undecidability of probabilistic planning and infinite-horizon partially observable Markov decision problems. In: Proc. of the 16th National Conference on Artificial Intelligence and 11th Conference on Innovative Applications of Artificial Intelligence, pp. 541–548. AAAI Press/The MIT Press, Menlo Park (1999)Google Scholar
- 18.Raz, R., Safra, S.: A sub-constant error-probability low-degree test, and a sub-constant error-probability PCP characterization of NP. In: Proc. of the 29th ACM Symposium on Theory of Computing, pp. 475–484 (1997)Google Scholar