Abstract
A general theory of intelligence must include learning, the
process of converting experiences into retrievable memories. We present
two CHREST models to illustrate the effects of learning across two different time scales (minutes and years, respectively). The first is an illustration of implicit learning, checking the validity of strings drawn from
an artificial grammar. The second looks at the interpretation of board-
game positions. The same learning and retrieval mechanisms are used
in both cases, and we argue that CHREST can be used by an artificial
general intelligence to construct and access declarative memory.
process of converting experiences into retrievable memories. We present
two CHREST models to illustrate the effects of learning across two different time scales (minutes and years, respectively). The first is an illustration of implicit learning, checking the validity of strings drawn from
an artificial grammar. The second looks at the interpretation of board-
game positions. The same learning and retrieval mechanisms are used
in both cases, and we argue that CHREST can be used by an artificial
general intelligence to construct and access declarative memory.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the Fifth Conference on Artificial General Intelligence |
| Place of Publication | Berlin, Heidelberg |
| Publisher | Springer Nature |
| Pages | 148-157 |
| Number of pages | 10 |
| Volume | LNAI 7716 |
| Publication status | Published - 2012 |