Another look at indirect negative evidence

Alexander Clark; Shalom Lappin

Another look at indirect negative evidence

Alexander Clark, Shalom Lappin

Research output: Contribution to conference › Paper › peer-review

Abstract

Indirect negative evidence is clearly an important way for learners to constrain overgeneralisation, and yet a good learning theoretic analysis has yet to be provided for this, whether in a PAC or a probabilistic identification in the limit framework. In this paper we suggest a theoretical analysis of indirect negative evidence that allows the presence of ungrammatical strings in the input and also accounts for the relationship between grammaticality/acceptability and probability. Given independently justified assumptions about lower bounds on the probabilities of grammatical strings, we establish that a limited number of membership queries of some strings can be probabilistically simulated.

Original language	English
Publication status	Published - 1 Mar 2009

Cite this

@conference{47aab5f3e1f94713bcd0d0c77983348a,

title = "Another look at indirect negative evidence",

abstract = "Indirect negative evidence is clearly an important way for learners to constrain overgeneralisation, and yet a good learning theoretic analysis has yet to be provided for this, whether in a PAC or a probabilistic identification in the limit framework. In this paper we suggest a theoretical analysis of indirect negative evidence that allows the presence of ungrammatical strings in the input and also accounts for the relationship between grammaticality/acceptability and probability. Given independently justified assumptions about lower bounds on the probabilities of grammatical strings, we establish that a limited number of membership queries of some strings can be probabilistically simulated.",

author = "Alexander Clark and Shalom Lappin",

year = "2009",

month = mar,

day = "1",

language = "English",

}

TY - CONF

T1 - Another look at indirect negative evidence

AU - Clark, Alexander

AU - Lappin, Shalom

PY - 2009/3/1

Y1 - 2009/3/1

N2 - Indirect negative evidence is clearly an important way for learners to constrain overgeneralisation, and yet a good learning theoretic analysis has yet to be provided for this, whether in a PAC or a probabilistic identification in the limit framework. In this paper we suggest a theoretical analysis of indirect negative evidence that allows the presence of ungrammatical strings in the input and also accounts for the relationship between grammaticality/acceptability and probability. Given independently justified assumptions about lower bounds on the probabilities of grammatical strings, we establish that a limited number of membership queries of some strings can be probabilistically simulated.

AB - Indirect negative evidence is clearly an important way for learners to constrain overgeneralisation, and yet a good learning theoretic analysis has yet to be provided for this, whether in a PAC or a probabilistic identification in the limit framework. In this paper we suggest a theoretical analysis of indirect negative evidence that allows the presence of ungrammatical strings in the input and also accounts for the relationship between grammaticality/acceptability and probability. Given independently justified assumptions about lower bounds on the probabilities of grammatical strings, we establish that a limited number of membership queries of some strings can be probabilistically simulated.

M3 - Paper

ER -