On learning formulas in the limit and with assurance

Andris Ambainis

doi:10.1016/S0020-0190(00)00150-2

On learning formulas in the limit and with assurance

Andris Ambainis

University of California at Berkeley

Research output: Contribution to journal › Article › peer-review

Abstract

The learning formulas in the model of Barzdins et al and Osherson et al were considered. It is shown that, in this model, a formula f can be learned in the limit if and only if not f can be learned with assurance. This explains some similarities between properties of limit learning and assurance learning.

Original language	English
Pages (from-to)	9-11
Number of pages	3
Journal	Information Processing Letters
Volume	77
Issue number	1
DOIs	https://doi.org/10.1016/S0020-0190(00)00150-2
Publication status	Published - 31 Jan 2001
Externally published	Yes

Access to Document

10.1016/S0020-0190(00)00150-2

Cite this

@article{f71c610440bb44d2a521efd1c95b03c6,

title = "On learning formulas in the limit and with assurance",

abstract = "The learning formulas in the model of Barzdins et al and Osherson et al were considered. It is shown that, in this model, a formula f can be learned in the limit if and only if not f can be learned with assurance. This explains some similarities between properties of limit learning and assurance learning.",

author = "Andris Ambainis",

year = "2001",

month = jan,

day = "31",

doi = "10.1016/S0020-0190(00)00150-2",

language = "English",

volume = "77",

pages = "9--11",

journal = "Information Processing Letters",

issn = "0020-0190",

publisher = "Elsevier B.V.",

number = "1",

}

TY - JOUR

T1 - On learning formulas in the limit and with assurance

AU - Ambainis, Andris

PY - 2001/1/31

Y1 - 2001/1/31

N2 - The learning formulas in the model of Barzdins et al and Osherson et al were considered. It is shown that, in this model, a formula f can be learned in the limit if and only if not f can be learned with assurance. This explains some similarities between properties of limit learning and assurance learning.

AB - The learning formulas in the model of Barzdins et al and Osherson et al were considered. It is shown that, in this model, a formula f can be learned in the limit if and only if not f can be learned with assurance. This explains some similarities between properties of limit learning and assurance learning.

UR - https://www.scopus.com/pages/publications/0035148280

U2 - 10.1016/S0020-0190(00)00150-2

DO - 10.1016/S0020-0190(00)00150-2

M3 - Article

AN - SCOPUS:0035148280

SN - 0020-0190

VL - 77

SP - 9

EP - 11

JO - Information Processing Letters

JF - Information Processing Letters

IS - 1

ER -

On learning formulas in the limit and with assurance

Abstract

Access to Document

Other files and links

Cite this