this paper can be generalized to the optimal model? 8 Acknowledgment I wish to thank L. Hemachandra for his invitation to me to spend the spring semester at the University of Rochester and for his permanent attention to my research and helpfulness in all my problems and J. Seiferas for extensive comments on an earlier draft of this paper. The results of section 4.1 of the paper are the realization of J. Seiferas's advice to investigate the probabilistic complexity properties of almost all functions in comparison with Yao's [Y1] results. I wish also to thank P. Dietz for his comments, which helped to simplify the proof of lemma 4.1

this paper contributes toward the goal of understanding how a computer can be programmed to learn by isolating features of incremental learning algorithms that theoretically enhance their learning potential. In particular, we examine the effects of imposing a limit on the amount of information that learning algorithm can hold in its memory as it attempts to This work was facilitated by an international agreement under NSF Grant 9119540.