A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf
.
Filters
On the learnability of recursively enumerable languages from good examples
2001
Theoretical Computer Science
The present paper investigates identiÿcation of indexed families L of recursively enumerable languages from good examples. ...
In the context of class-preserving learning, it is shown that the learning power of ÿnite and limit identiÿcation from good text examples coincide. ...
We show that there are indexable classes of r.e. languages which are class preservingly learnable from good examples, but which are not learnable in the limit from text. ...
doi:10.1016/s0304-3975(00)00131-6
fatcat:arn4dtjv7fggxchneffbj7pzjm
One-Shot Learners Using Negative Counterexamples and Nearest Positive Examples
[chapter]
2007
Lecture Notes in Computer Science
We compare and explore general capabilities of our models, as well as complexity advantages of learnability models of one type over models of other types, where complexity is measured in terms of number ...
In particular, we establish that "correcting" positive examples are sometimes more helpful to a learner than just negative (counter) examples and access to full positive data. ...
Acknowledgements We thank the anonymous referees for several helpful comments which improved the presentation of this paper. ...
doi:10.1007/978-3-540-75225-7_22
fatcat:kfzjndxiqrgbtjpqnxg6jnpg24
One-shot learners using negative counterexamples and nearest positive examples
2009
Theoretical Computer Science
We compare and explore general capabilities of our models, as well as complexity advantages of learnability models of one type over models of other types, where complexity is measured in terms of number ...
In particular, we establish that "correcting" positive examples are sometimes more helpful to a learner than just negative (counter) examples and access to full positive data. ...
Acknowledgements We thank the anonymous referees for several helpful comments which improved the presentation of this paper. ...
doi:10.1016/j.tcs.2009.02.013
fatcat:elooedvfa5fedpiw3ulazem3si
Page 4188 of Mathematical Reviews Vol. , Issue 2001F
[page]
2001
Mathematical Reviews
As it will turn out, there are enumerable classes of recursive functions that are not learnable from ‘typical’ examples. The learnable classes are characterized. ...
{For the entire collection see MR 2001d:91002.}
2001f:68058 68Q32
Nessel, Jochen (D-KSRL; Kaiserslautern)
Learnability of enumerable classes of recursive functions from “typical” examples. ...
Self-Organizing Machine Translation: Example-Driven Induction of Transfer Functions
[article]
1994
arXiv
pre-print
With the advent of faster computers, the notion of doing machine translation from a huge stored database of translation examples is no longer unreasonable. ...
This system has been used to infer English->French and English->Urdu transfer functions from small corpora. ...
I would like to thank Jerry Feldman of ICSI for the opportunity to present a preliminary version of these results to his team, and Andreas Stolcke, Jonathan Segal, and Dan Jurafsky for their feedback from ...
arXiv:cmp-lg/9406012v1
fatcat:eeqyzule2vaz3eaeitfikmmk2q
Partial learning of recursively enumerable languages
2016
Theoretical Computer Science
In particular a class is consistently partially learnable iff it is a subclass of a uniformly recursive family. F. ...
This paper studies several typical learning criteria in the model of partial learning of r.e. sets in the recursion-theoretic framework of inductive inference. ...
Gao, Jain and Stephan [7, Example 9] have proven that the class of graphs of all recursive functions witnesses this separation. Theorem 27 (Gao, Jain and Stephan [7] ). ...
doi:10.1016/j.tcs.2015.10.035
fatcat:4d52ygubgnd5hkskxlbrgna7tq
Partial Learning of Recursively Enumerable Languages
[chapter]
2013
Lecture Notes in Computer Science
In particular a class is consistently partially learnable iff it is a subclass of a uniformly recursive family. F. ...
This paper studies several typical learning criteria in the model of partial learning of r.e. sets in the recursion-theoretic framework of inductive inference. ...
Gao, Jain and Stephan [7, Example 9] have proven that the class of graphs of all recursive functions witnesses this separation. Theorem 27 (Gao, Jain and Stephan [7] ). ...
doi:10.1007/978-3-642-40935-6_9
fatcat:pdybl5ypurccrhzzlohihzcncq
Learning Simple Concepts under Simple Distributions
1991
SIAM journal on computing (Print)
Using the definition of µ T , it is easy to define a recursive function which approximates µ T from below in the sense of (11). Hence µ T is an enumerable measure. ...
We extend the notion of prefix complexity to recursively enumerable sets and enumerable functions. For each set D ⊆ N define: m(D) = Σ {m(y): y ∈ D} K (D) = − log m(D) + O (1). ...
doi:10.1137/0220056
fatcat:kmyydxxkfrfhjbukltykwlihje
Learning indexed families of recursive languages from positive data: A survey
2008
Theoretical Computer Science
This model has been studied for learning recursive functions, recursively enumerable languages, and recursive languages, reflecting different aspects of machine learning, artificial intelligence, complexity ...
One line of research focuses on indexed families of recursive languages -classes of recursive languages described in a representation scheme for which the question of membership for any string in any of ...
Telltales are also relevant for learning recursive functions from examples. ...
doi:10.1016/j.tcs.2008.02.030
fatcat:jzsqdyddvrbz3bhwxeukqnrty4
Robust Learning Aided by Context
2000
Journal of computer and system sciences (Print)
The class REC of all computable functions is robustly learnable from a selected context. ...
In this paper they also showed that if a class is closed under finite variants and robustly learnable, then it is contained in a recursively enumerable class. ...
doi:10.1006/jcss.1999.1637
fatcat:wdbztuag5jfdjnbg6yikev6hoq
Robust learning aided by context
1998
Proceedings of the eleventh annual conference on Computational learning theory - COLT' 98
Also, studied is the difficulty of the functional dependence between the intended target tasks and useful associated contexts. ...
Empirical studies of multitask learning provide some evidence that the performance of a learning system on its intended targets improves by presenting to the learning system related tasks, also called ...
The class REC of all computable functions is robustly learnable from selected context. ...
doi:10.1145/279943.279952
dblp:conf/colt/CaseJOSS98
fatcat:nhok3hto5feuxbqzbo3hunfqay
Page 5 of Mathematical Reviews Vol. 53, Issue 4
[page]
1977
Mathematical Reviews
From the authors’ summary: “The axiomatic treatment of the computational complexity of partial recursive functions initiated by Blum is extended to relatively computable functions (as computed, for example ...
The second main result concerns the classes (€"), consisting of predicates (i.c., 0-1 functions) from &". ...
Invertible Classes
[chapter]
2006
Lecture Notes in Computer Science
This paper considers when one can invert general recursive operators which map a class of functions F to F . In this regard, we study four different notions of inversion. ...
Three different possible types of enumeration are studied. ...
Typical examples from the more applied world are the following ones: -Cryptography. ...
doi:10.1007/11750321_67
fatcat:p7msuvau4valplnnnmei2k24te
Invertible classes
2007
Theoretical Computer Science
This paper considers when one can invert general recursive operators which map a class of functions F to F . In this regard, we study four different notions of inversion. ...
Three different possible types of enumeration are studied. ...
Typical examples from the more applied world are the following ones: -Cryptography. ...
doi:10.1016/j.tcs.2007.05.017
fatcat:5cbaqqyn4jhfxoplzoeimaikuq
The Synthesis of Language Learners
1999
Information and Computation
For example, the indexed families of recursive languages that can be behaviorally correctly identified from positive data are surprisingly characterized by Angluin's condition 2 (the subset principle for ...
Studied is the metaproblem of synthesizing from indices for r.e. classes and for indexed families of languages various kinds of language learners for the corresponding classes or families indexed. ...
Furthermore, the proofs of the positive results provide characterizations of corresponding noise-tolerantly learnable language classes. ...
doi:10.1006/inco.1998.2782
fatcat:bkrvheyrybgtlpg2d5azxxoh7q
« Previous
Showing results 1 — 15 out of 862 results