A survey of some aspects of computational learning theory

摘要: In the introduction of his paper starting computational learning theory, Valiant observed that intuitive notion merits similar attention from point view formal theoretical study as computing. this comparison, appears to be more elusive, difficult capture by a unified mathematical theory (as noted Haussler (1990), it is not clear whether such even possible or desirable). Research was focused on concept learning, which in fact closely related computing several approaches developed computer science can adapted its study. Interesting connections were found with other fields combinatorial optimization, cryptography and statistical pattern recognition. survey we gave short account some aspects results obtained describing models, characterizations learnability, algorithms negative results.