Keywords:representaton, model, homomorphism
Statistical learning acquire knowledge from the environment based on counting the number of occurrences of events. Thus, flexibly defining a set of events contributes to an expanded range of reasoning. In this paper, we first define the alignment structure, which is an essential representation for every statistical learning. This consists of a case index set (CIS) having an index for specifying an event as an element, and a value reading procedure for reading a value corresponding to each variable from a value area according to each index. The basic procedure of constructing an abstract CIS is repetition of process that is a procedure that cuts out a partial area starting from a certain element in the CIS and associates it with a higher-order CIS element. This consideration is expected to contribute to the research of artificial intelligence in the future, which will expand intelligence by automatically finding alignment structures.
Authentication for paper PDF access
A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.