LEARNING VIA QUERIES IN [+, &lt;]

Gasarch, William; Pleszkoch, Mark G.; Solovay, Robert M.

doi:10.1016/b978-1-55860-146-8.50029-1

Cited by 7 publications

(18 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By examining the complexity of queries made by query inference machines, we show that separation results for many standard inference types remain true for their query counterparts. This settles a number of conjectures in query inference [17,19].…”

Section: Introductionmentioning

confidence: 89%

“…We then demonstrate the incomparability by considering the extreme cases (by comparing answer inference types with standard inference types such as PEX 0 and [1, n] QBC a (L)). Another theme of this paper is to demonstrate how answer inference types can be used as a technical tool in the study of query inference [19]. By examining the complexity of queries made by query inference machines, we show that separation results for many standard inference types remain true for their query counterparts.…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Inferring answers to queries

Gasarch

Lee

2008

Journal of Computer and System Sciences

Self Cite

View full text Add to dashboard Cite

One focus of inductive inference is to infer a program for a function f from observations or queries about f . We propose a new line of research which examines the question of inferring the answers to queries. For a given class of computable functions, we consider the learning (in the limit) of properties of these functions that can be captured by queries formulated in a logical language L. We study the inference types that arise in this context. Of particular interest is a comparison between the learning of properties and the learning of programs. Our results suggest that these two types of learning are incomparable. In addition, our techniques can be used to prove a general lemma about query inference [W. Gasarch, C. Smith, Learning via queries, J. ACM 39 (1992) 649-676]. We show that I ⊂ J ⇒ QI(L) ⊂ QJ (L) for many standard inference types I, J and many query languages L. Hence any separation that holds between these inference types also holds between the corresponding query inference types. One interesting consequence is that [24, 49] QEX 0 [Succ, <] 2 − [2, 4] QEX 0 [Succ, <] 2 = ∅.Published by Elsevier Inc.

show abstract

Section: Introductionmentioning

confidence: 89%

Section: Introductionmentioning

confidence: 98%

Inferring answers to queries

Gasarch

Lee

2008

Journal of Computer and System Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition to this, we also show that we can trade mind changes for additional members without altering the learning power only if we are dealing with errors of commission. We now describe a model of learning (see [8,16,17,19,20,26]) that will make the above mentioned result more formal. In the case of team learning with success rate being m out of n, the team is considered to be successful in learning a target function f if at least m members of the team produce programs that compute f correctly.…”

Section: Introductionmentioning

confidence: 96%

“…Although counting the number of mind changes an IIM makes before converging is not an abstract measure of the complexity of inference [12], it does provide a reasonable estimate for implemented inference systems. Consequently, the number of mind changes made by inference machines has received considerable attention [6,7,9,14,15,20,23,25,26]. A subscript b on the class name indicates a success criterion where the IIM converges after no more than b changes of conjecture.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Taming teams with mind changes

Kalyanasundaram¹,

Velauthapillai²

2008

Journal of Computer and System Sciences

View full text Add to dashboard Cite

This paper is concerned with the algorithmic learning where the learner is allowed to make a finite but bounded number of mind changes. Briefly, in our learning paradigm, a learner is given examples from a recursive function, which the learner attempts to learn by producing programs to compute that function. We say that a team is successful if at least one member of the team learns the target function. The problem, given two teams with bounded number of mind changes whether, one team can provably learn more than the other team, was first proposed by Smith [C.H. Smith, The power of pluralism for automatic program synthesis, J. Assoc. Comput. Mach. 29 (1982Mach. 29 ( ) 1144Mach. 29 ( -1165. This problem has been open for the last twenty five years. This paper makes progress toward a complete solution of this problem. In the case of error-free learning, this paper closes the gap between the lower and the upper bounds. Finally, in the case of EX learning our result shows that there is no team with a 0 mind changes whose learning power is exactly equal to a single learner with bounded b ( = a) number of mind changes. In the case of Popperian learning (PEX) we have a positive answer.

show abstract

Remarks on intelligence as extended retrieval and its implications

Chen

1992

AI & Soc

View full text Add to dashboard Cite

Abstract. The current understanding of the important role of information or knowledge retrieval to artificial intelligence is not adequate. In this article, three study areas related to retrieval are summarized and commented. The need for a more systematic and unified way of studying retrieval and intelligence is argued. A brief outline for future study is suggested, some philosophical and social implications of this study are also discussed.

show abstract

LEARNING VIA QUERIES IN [+, <]

Cited by 7 publications

References 5 publications

Inferring answers to queries

Inferring answers to queries

Taming teams with mind changes

Remarks on intelligence as extended retrieval and its implications

Contact Info

Product

Resources

About