Abstract
Summary form only given, as follows. Inference of predicate logic, which is widely used for representation of knowledge in artificial intelligence (AI) systems, is discussed from the viewpoints of source coding and decision theory. Since inference in logic can be regarded as some kind of information transformation, an analogy between inference and source coding can be observed. Inductive inference is regarded as source encoding, because observed facts or examples are compressed into an axiom similar to the compression of a source sequence into a code word. On the other hand, deductive inference is interpreted as decoding. From the viewpoint of decision theory, inductive inference is regarded as the decision problem selecting the collect axiom which represents an observing world. A new inductive inference scheme which induces the minimum Bayes risk is proposed. A method for selecting the axiom which represents the finite observed facts by the minimum description length code is shown.
| Original language | English |
|---|---|
| Title of host publication | 1990 IEEE Int Symp Inf Theor |
| Place of Publication | Piscataway, NJ, United States |
| Publisher | Publ by IEEE |
| Pages | 101 |
| Number of pages | 1 |
| Publication status | Published - 1990 |
| Externally published | Yes |
| Event | 1990 IEEE International Symposium on Information Theory - San Diego, CA, USA Duration: 1990 Jan 14 → 1990 Jan 19 |
Other
| Other | 1990 IEEE International Symposium on Information Theory |
|---|---|
| City | San Diego, CA, USA |
| Period | 90/1/14 → 90/1/19 |
Fingerprint
ASJC Scopus subject areas
- Engineering(all)
Cite this
On the optimal inductive inference scheme from the view point of source coding. / Matsushima, Toshiyasu; Suzuki, Joe; Inazumi, Hiroshige; Hirasawa, Shigeichi.
1990 IEEE Int Symp Inf Theor. Piscataway, NJ, United States : Publ by IEEE, 1990. p. 101.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - On the optimal inductive inference scheme from the view point of source coding
AU - Matsushima, Toshiyasu
AU - Suzuki, Joe
AU - Inazumi, Hiroshige
AU - Hirasawa, Shigeichi
PY - 1990
Y1 - 1990
N2 - Summary form only given, as follows. Inference of predicate logic, which is widely used for representation of knowledge in artificial intelligence (AI) systems, is discussed from the viewpoints of source coding and decision theory. Since inference in logic can be regarded as some kind of information transformation, an analogy between inference and source coding can be observed. Inductive inference is regarded as source encoding, because observed facts or examples are compressed into an axiom similar to the compression of a source sequence into a code word. On the other hand, deductive inference is interpreted as decoding. From the viewpoint of decision theory, inductive inference is regarded as the decision problem selecting the collect axiom which represents an observing world. A new inductive inference scheme which induces the minimum Bayes risk is proposed. A method for selecting the axiom which represents the finite observed facts by the minimum description length code is shown.
AB - Summary form only given, as follows. Inference of predicate logic, which is widely used for representation of knowledge in artificial intelligence (AI) systems, is discussed from the viewpoints of source coding and decision theory. Since inference in logic can be regarded as some kind of information transformation, an analogy between inference and source coding can be observed. Inductive inference is regarded as source encoding, because observed facts or examples are compressed into an axiom similar to the compression of a source sequence into a code word. On the other hand, deductive inference is interpreted as decoding. From the viewpoint of decision theory, inductive inference is regarded as the decision problem selecting the collect axiom which represents an observing world. A new inductive inference scheme which induces the minimum Bayes risk is proposed. A method for selecting the axiom which represents the finite observed facts by the minimum description length code is shown.
UR - http://www.scopus.com/inward/record.url?scp=0025628156&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0025628156&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:0025628156
SP - 101
BT - 1990 IEEE Int Symp Inf Theor
PB - Publ by IEEE
CY - Piscataway, NJ, United States
ER -