skip to content
Active Learning Using Arbitrary Binary Valued Queries Preview this item
ClosePreview this item
Checking...

Active Learning Using Arbitrary Binary Valued Queries

Author: S R KulKarni; S K Mitter; J N Tsitsiklis; MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS.
Publisher: Ft. Belvoir Defense Technical Information Center OCT 1990.
Edition/Format:   eBook : EnglishView all editions and formats
Database:WorldCat
Summary:
The original and most widely studied PAC model for learning assumes a passive learner in the sense that the learner plays no role in obtaining information about the unknown concept. That is, the samples are simply drawn independently from some probability distribution. Some work has been done on studying more powerful oracles and how they affect learnability. To find bounds on the improvement that can be expected  Read more...
Rating:

(not yet rated) 0 with reviews - Be the first.

Subjects
More like this

 

Find a copy online

Links to this item

Find a copy in the library

&AllPage.SpinnerRetrieving; Finding libraries that hold this item...

Details

Material Type: Internet resource
Document Type: Internet Resource
All Authors / Contributors: S R KulKarni; S K Mitter; J N Tsitsiklis; MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS.
OCLC Number: 227911837
Notes: Sponsored in part by the National Science Foundation under Grant No., ECS-8552419.
Description: 13 pages

Abstract:

The original and most widely studied PAC model for learning assumes a passive learner in the sense that the learner plays no role in obtaining information about the unknown concept. That is, the samples are simply drawn independently from some probability distribution. Some work has been done on studying more powerful oracles and how they affect learnability. To find bounds on the improvement that can be expected from using oracles, we consider active learning in the sense that the learner has complete choice in the information received. Specifically, we allow the learner to ask arbitrary yes/no questions. We consider both active learning under a fixed distribution and distribution-free active learning. In the case of active learning, the underlying probability distribution is used only to measure distance between concepts. For learnability with respect to a fixed distribution, active learning does not enlarge the set of learnable concept classes, but can improve the sample complexity. For distribution-free learning, it is shown that a concept class is actively learnable iff it is finite, so that active learning is in fact less powerful than the usual passive learning model. We also consider a form of distribution-free learning in which the learner knows the distribution being used, so that 'distribution-free' refers only to the requirement that a bound on the number of queries can be obtained uniformly over all distributions. Even with the side information of the distribution being used, a concept class is actively learnable iff it has finite VC dimension, so that active learning with the side information still does not enlarge the set of learnable concept classes.

Reviews

User-contributed reviews
Retrieving GoodReads reviews...
Retrieving DOGObooks reviews...

Tags

Be the first.
Confirm this request

You may have already requested this item. Please select Ok if you would like to proceed with this request anyway.

Linked Data


Primary Entity

<http://www.worldcat.org/oclc/227911837> # Active Learning Using Arbitrary Binary Valued Queries
    a schema:Book, schema:CreativeWork ;
   library:oclcnum "227911837" ;
   library:placeOfPublication <http://id.loc.gov/vocabulary/countries/vau> ;
   library:placeOfPublication <http://experiment.worldcat.org/entity/work/data/24094705#Place/ft_belvoir> ; # Ft. Belvoir
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Thing/active_learning> ; # ACTIVE LEARNING
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/sampling> ; # Sampling
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/inverse_problems> ; # Inverse problems
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/probability_distribution_functions> ; # Probability distribution functions
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/statistics_and_probability> ; # Statistics and Probability
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Thing/oracles> ; # ORACLES
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/models> ; # Models
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/psychology> ; # Psychology
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/learning> ; # Learning
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Thing/distribution_free_learning> ; # DISTRIBUTION-FREE LEARNING
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/arbitration> ; # Arbitration
   schema:about <http://experiment.worldcat.org/entity/work/data/24094705#Topic/queueing_theory> ; # Queueing theory
   schema:bookFormat schema:EBook ;
   schema:contributor <http://viaf.org/viaf/218256027> ; # J. N. Tsitsiklis
   schema:contributor <http://viaf.org/viaf/311183078> ; # S. K. Mitter
   schema:contributor <http://experiment.worldcat.org/entity/work/data/24094705#Person/kulkarni_s_r> ; # S. R. KulKarni
   schema:contributor <http://experiment.worldcat.org/entity/work/data/24094705#Organization/massachusetts_inst_of_tech_cambridge_lab_for_information_and_decision_systems> ; # MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS.
   schema:datePublished "OCT 1990" ;
   schema:datePublished "1990" ;
   schema:description "The original and most widely studied PAC model for learning assumes a passive learner in the sense that the learner plays no role in obtaining information about the unknown concept. That is, the samples are simply drawn independently from some probability distribution. Some work has been done on studying more powerful oracles and how they affect learnability. To find bounds on the improvement that can be expected from using oracles, we consider active learning in the sense that the learner has complete choice in the information received. Specifically, we allow the learner to ask arbitrary yes/no questions. We consider both active learning under a fixed distribution and distribution-free active learning. In the case of active learning, the underlying probability distribution is used only to measure distance between concepts. For learnability with respect to a fixed distribution, active learning does not enlarge the set of learnable concept classes, but can improve the sample complexity. For distribution-free learning, it is shown that a concept class is actively learnable iff it is finite, so that active learning is in fact less powerful than the usual passive learning model. We also consider a form of distribution-free learning in which the learner knows the distribution being used, so that 'distribution-free' refers only to the requirement that a bound on the number of queries can be obtained uniformly over all distributions. Even with the side information of the distribution being used, a concept class is actively learnable iff it has finite VC dimension, so that active learning with the side information still does not enlarge the set of learnable concept classes."@en ;
   schema:exampleOfWork <http://worldcat.org/entity/work/id/24094705> ;
   schema:inLanguage "en" ;
   schema:name "Active Learning Using Arbitrary Binary Valued Queries"@en ;
   schema:productID "227911837" ;
   schema:publication <http://www.worldcat.org/title/-/oclc/227911837#PublicationEvent/ft_belvoirdefense_technical_information_centeroct_1990> ;
   schema:publisher <http://experiment.worldcat.org/entity/work/data/24094705#Agent/defense_technical_information_center> ; # Defense Technical Information Center
   schema:url <http://handle.dtic.mil/100.2/ADA459598> ;
   wdrs:describedby <http://www.worldcat.org/title/-/oclc/227911837> ;
    .


Related Entities

<http://experiment.worldcat.org/entity/work/data/24094705#Agent/defense_technical_information_center> # Defense Technical Information Center
    a bgn:Agent ;
   schema:name "Defense Technical Information Center" ;
    .

<http://experiment.worldcat.org/entity/work/data/24094705#Organization/massachusetts_inst_of_tech_cambridge_lab_for_information_and_decision_systems> # MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS.
    a schema:Organization ;
   schema:name "MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS." ;
    .

<http://experiment.worldcat.org/entity/work/data/24094705#Person/kulkarni_s_r> # S. R. KulKarni
    a schema:Person ;
   schema:familyName "KulKarni" ;
   schema:givenName "S. R." ;
   schema:name "S. R. KulKarni" ;
    .

<http://experiment.worldcat.org/entity/work/data/24094705#Thing/distribution_free_learning> # DISTRIBUTION-FREE LEARNING
    a schema:Thing ;
   schema:name "DISTRIBUTION-FREE LEARNING" ;
    .

<http://experiment.worldcat.org/entity/work/data/24094705#Topic/probability_distribution_functions> # Probability distribution functions
    a schema:Intangible ;
   schema:name "Probability distribution functions"@en ;
    .

<http://experiment.worldcat.org/entity/work/data/24094705#Topic/statistics_and_probability> # Statistics and Probability
    a schema:Intangible ;
   schema:name "Statistics and Probability"@en ;
    .

<http://viaf.org/viaf/218256027> # J. N. Tsitsiklis
    a schema:Person ;
   schema:familyName "Tsitsiklis" ;
   schema:givenName "J. N." ;
   schema:name "J. N. Tsitsiklis" ;
    .

<http://viaf.org/viaf/311183078> # S. K. Mitter
    a schema:Person ;
   schema:familyName "Mitter" ;
   schema:givenName "S. K." ;
   schema:name "S. K. Mitter" ;
    .


Content-negotiable representations

Close Window

Please sign in to WorldCat 

Don't have an account? You can easily create a free account.