Learning with a Limited Memory

Rusins Freivalds and Efim Kinber and Carl Smith

The invention of the formal study of inductive inference was a direct consequence of efforts to moclel human learning [Go167a, Sch63a]. Much of the early work in inductive inference [Ang83a, Kle80a, Osh86a] and the more recent work [Hau88a, RivS9a, Ful90a, Nat91a, War91a, Da192a] has been motivated by trying to model or understand some aspect of human learning in the hopes that the knowledge gained may be useful in the effort to program a computer to learn. For example, one of the intuitions behind Valiant’s PAC model [Val84a] was that humans learn quickly. This paper is motivated by the intuition that people forget.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.