Universal parameter optimisation in games based on SPSA View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2006-06

AUTHORS

Levente Kocsis, Csaba Szepesvári

ABSTRACT

Most game programs have a large number of parameters that are crucial for their performance. While tuning these parameters by hand is rather difficult, efficient and easy to use generic automatic parameter optimisation algorithms are known only for special problems such as the adjustment of the parameters of an evaluation function. The SPSA algorithm (Simultaneous Perturbation Stochastic Approximation) is a generic stochastic gradient method for optimising an objective function when an analytic expression of the gradient is not available, a frequent case in game programs. Further, SPSA in its canonical form is very easy to implement. As such, it is an attractive choice for parameter optimisation in game programs, both due to its generality and simplicity. The goal of this paper is twofold: (i) to introduce SPSA for the game programming community by putting it into a game-programming perspective, and (ii) to propose and discuss several methods that can be used to enhance the performance of SPSA. These methods include using common random numbers and antithetic variables, a combination of SPSA with RPROP, and the reuse of samples of previous performance evaluations. SPSA with the proposed enhancements was tested in some large-scale experiments on tuning the parameters of an opponent model, a policy and an evaluation function in our poker program, MCRAISE. Whilst SPSA with no enhancements failed to make progress using the allocated resources, SPSA with the enhancements proved to be competitive with other methods, including TD-learning; increasing the average payoff per game by as large as 0.19 times the size of the amount of the small bet. From the experimental study, we conclude that the use of an appropriately enhanced variant of SPSA for the optimisation of game program parameters is a viable approach, especially if no good alternative exist for the types of parameters considered. More... »

PAGES

249-286

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s10994-006-6888-8

DOI

http://dx.doi.org/10.1007/s10994-006-6888-8

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1012359325


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "MTA Institute for Computer Science and Control", 
          "id": "https://www.grid.ac/institutes/grid.4836.9", 
          "name": [
            "MTA SZTAKI, Kende u. 13\u201317, 1111, Budapest, Hungary"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kocsis", 
        "givenName": "Levente", 
        "id": "sg:person.07770302215.82", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07770302215.82"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "MTA Institute for Computer Science and Control", 
          "id": "https://www.grid.ac/institutes/grid.4836.9", 
          "name": [
            "MTA SZTAKI, Kende u. 13\u201317, 1111, Budapest, Hungary"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Szepesv\u00e1ri", 
        "givenName": "Csaba", 
        "id": "sg:person.016202177221.23", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016202177221.23"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/bf00992696", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1001381236", 
          "https://doi.org/10.1007/bf00992696"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1019636471", 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4899-2696-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1019636471", 
          "https://doi.org/10.1007/978-1-4899-2696-8"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4899-2696-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1019636471", 
          "https://doi.org/10.1007/978-1-4899-2696-8"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.neucom.2004.11.016", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1023604540"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf00115009", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1028679285", 
          "https://doi.org/10.1007/bf00115009"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf00992697", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1029271228", 
          "https://doi.org/10.1007/bf00992697"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0004-3702(01)00130-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1044471374"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1023/a:1007634325138", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1047698138", 
          "https://doi.org/10.1023/a:1007634325138"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0020-0255(03)00045-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1049691245"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0020-0255(03)00045-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1049691245"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0925-2312(01)00700-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1052928518"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/72.809083", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061219348"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/9.119632", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061242810"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tac.2000.880982", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061474938"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tac.2003.815008", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061475376"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tnn.1998.712192", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061716400"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1137/s1052623495270723", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1062883483"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aoms/1177698258", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064399247"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aoms/1177728659", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064401477"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aoms/1177729392", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064401671"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aoms/1177729586", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064401743"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aos/1059655913", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064406449"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1214/aos/1176350965", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064409307"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1287/mnsc.31.1.66", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064719947"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1287/mnsc.38.6.884", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064720955"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1287/mnsc.45.11.1570", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064721842"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/icnn.1993.298623", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1086259851"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/iconip.2002.1198180", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093470523"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/ispa.2005.195385", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1093503311"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/acc.1997.609490", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094523266"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/wsc.1998.744910", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095505241"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/wsc.2002.1172897", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095737253"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1049/cp:19991170", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1098689874"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1613/jair.806", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1105579523"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/0471722138", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1109491948"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/0471722138", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1109491948"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2006-06", 
    "datePublishedReg": "2006-06-01", 
    "description": "Most game programs have a large number of parameters that are crucial for their performance. While tuning these parameters by hand is rather difficult, efficient and easy to use generic automatic parameter optimisation algorithms are known only for special problems such as the adjustment of the parameters of an evaluation function. The SPSA algorithm (Simultaneous Perturbation Stochastic Approximation) is a generic stochastic gradient method for optimising an objective function when an analytic expression of the gradient is not available, a frequent case in game programs. Further, SPSA in its canonical form is very easy to implement. As such, it is an attractive choice for parameter optimisation in game programs, both due to its generality and simplicity. The goal of this paper is twofold: (i) to introduce SPSA for the game programming community by putting it into a game-programming perspective, and (ii) to propose and discuss several methods that can be used to enhance the performance of SPSA. These methods include using common random numbers and antithetic variables, a combination of SPSA with RPROP, and the reuse of samples of previous performance evaluations. SPSA with the proposed enhancements was tested in some large-scale experiments on tuning the parameters of an opponent model, a policy and an evaluation function in our poker program, MCRAISE. Whilst SPSA with no enhancements failed to make progress using the allocated resources, SPSA with the enhancements proved to be competitive with other methods, including TD-learning; increasing the average payoff per game by as large as 0.19 times the size of the amount of the small bet. From the experimental study, we conclude that the use of an appropriately enhanced variant of SPSA for the optimisation of game program parameters is a viable approach, especially if no good alternative exist for the types of parameters considered.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1007/s10994-006-6888-8", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1125588", 
        "issn": [
          "0885-6125", 
          "1573-0565"
        ], 
        "name": "Machine Learning", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "3", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "63"
      }
    ], 
    "name": "Universal parameter optimisation in games based on SPSA", 
    "pagination": "249-286", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "3ae256a49d9e9c6b3d5cc3ab4fab14c5c35d9e8c5e1e7d78ae481839e3fed5d1"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s10994-006-6888-8"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1012359325"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s10994-006-6888-8", 
      "https://app.dimensions.ai/details/publication/pub.1012359325"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T12:58", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000365_0000000365/records_71683_00000000.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "http://link.springer.com/10.1007%2Fs10994-006-6888-8"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s10994-006-6888-8'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s10994-006-6888-8'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s10994-006-6888-8'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s10994-006-6888-8'


 

This table displays all metadata directly associated to this object as RDF triples.

174 TRIPLES      21 PREDICATES      61 URIs      19 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s10994-006-6888-8 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author Na878c01f57be41ebb15e93a9650620cc
4 schema:citation sg:pub.10.1007/978-1-4899-2696-8
5 sg:pub.10.1007/bf00115009
6 sg:pub.10.1007/bf00992696
7 sg:pub.10.1007/bf00992697
8 sg:pub.10.1023/a:1007634325138
9 https://app.dimensions.ai/details/publication/pub.1019636471
10 https://doi.org/10.1002/0471722138
11 https://doi.org/10.1016/j.neucom.2004.11.016
12 https://doi.org/10.1016/s0004-3702(01)00130-8
13 https://doi.org/10.1016/s0020-0255(03)00045-8
14 https://doi.org/10.1016/s0925-2312(01)00700-7
15 https://doi.org/10.1049/cp:19991170
16 https://doi.org/10.1109/72.809083
17 https://doi.org/10.1109/9.119632
18 https://doi.org/10.1109/acc.1997.609490
19 https://doi.org/10.1109/icnn.1993.298623
20 https://doi.org/10.1109/iconip.2002.1198180
21 https://doi.org/10.1109/ispa.2005.195385
22 https://doi.org/10.1109/tac.2000.880982
23 https://doi.org/10.1109/tac.2003.815008
24 https://doi.org/10.1109/tnn.1998.712192
25 https://doi.org/10.1109/wsc.1998.744910
26 https://doi.org/10.1109/wsc.2002.1172897
27 https://doi.org/10.1137/s1052623495270723
28 https://doi.org/10.1214/aoms/1177698258
29 https://doi.org/10.1214/aoms/1177728659
30 https://doi.org/10.1214/aoms/1177729392
31 https://doi.org/10.1214/aoms/1177729586
32 https://doi.org/10.1214/aos/1059655913
33 https://doi.org/10.1214/aos/1176350965
34 https://doi.org/10.1287/mnsc.31.1.66
35 https://doi.org/10.1287/mnsc.38.6.884
36 https://doi.org/10.1287/mnsc.45.11.1570
37 https://doi.org/10.1613/jair.806
38 schema:datePublished 2006-06
39 schema:datePublishedReg 2006-06-01
40 schema:description Most game programs have a large number of parameters that are crucial for their performance. While tuning these parameters by hand is rather difficult, efficient and easy to use generic automatic parameter optimisation algorithms are known only for special problems such as the adjustment of the parameters of an evaluation function. The SPSA algorithm (Simultaneous Perturbation Stochastic Approximation) is a generic stochastic gradient method for optimising an objective function when an analytic expression of the gradient is not available, a frequent case in game programs. Further, SPSA in its canonical form is very easy to implement. As such, it is an attractive choice for parameter optimisation in game programs, both due to its generality and simplicity. The goal of this paper is twofold: (i) to introduce SPSA for the game programming community by putting it into a game-programming perspective, and (ii) to propose and discuss several methods that can be used to enhance the performance of SPSA. These methods include using common random numbers and antithetic variables, a combination of SPSA with RPROP, and the reuse of samples of previous performance evaluations. SPSA with the proposed enhancements was tested in some large-scale experiments on tuning the parameters of an opponent model, a policy and an evaluation function in our poker program, MCRAISE. Whilst SPSA with no enhancements failed to make progress using the allocated resources, SPSA with the enhancements proved to be competitive with other methods, including TD-learning; increasing the average payoff per game by as large as 0.19 times the size of the amount of the small bet. From the experimental study, we conclude that the use of an appropriately enhanced variant of SPSA for the optimisation of game program parameters is a viable approach, especially if no good alternative exist for the types of parameters considered.
41 schema:genre research_article
42 schema:inLanguage en
43 schema:isAccessibleForFree true
44 schema:isPartOf N72dacd96de9343999fe61b42e73e62e8
45 Na8700225396346fd8b6f1723df143952
46 sg:journal.1125588
47 schema:name Universal parameter optimisation in games based on SPSA
48 schema:pagination 249-286
49 schema:productId N5bd3f9a1340c428ea71cf1fe38672a8e
50 N618fe067d1c7492088585055c19d10ce
51 Nac1f809726cc41fd984f172b0eb009aa
52 schema:sameAs https://app.dimensions.ai/details/publication/pub.1012359325
53 https://doi.org/10.1007/s10994-006-6888-8
54 schema:sdDatePublished 2019-04-11T12:58
55 schema:sdLicense https://scigraph.springernature.com/explorer/license/
56 schema:sdPublisher N4b0d10ce908c481685b8f580a62c3e1f
57 schema:url http://link.springer.com/10.1007%2Fs10994-006-6888-8
58 sgo:license sg:explorer/license/
59 sgo:sdDataset articles
60 rdf:type schema:ScholarlyArticle
61 N4b0d10ce908c481685b8f580a62c3e1f schema:name Springer Nature - SN SciGraph project
62 rdf:type schema:Organization
63 N5bd3f9a1340c428ea71cf1fe38672a8e schema:name doi
64 schema:value 10.1007/s10994-006-6888-8
65 rdf:type schema:PropertyValue
66 N618fe067d1c7492088585055c19d10ce schema:name readcube_id
67 schema:value 3ae256a49d9e9c6b3d5cc3ab4fab14c5c35d9e8c5e1e7d78ae481839e3fed5d1
68 rdf:type schema:PropertyValue
69 N72dacd96de9343999fe61b42e73e62e8 schema:volumeNumber 63
70 rdf:type schema:PublicationVolume
71 Na8700225396346fd8b6f1723df143952 schema:issueNumber 3
72 rdf:type schema:PublicationIssue
73 Na878c01f57be41ebb15e93a9650620cc rdf:first sg:person.07770302215.82
74 rdf:rest Ndfbe1ba03b164c639f88e9ecc1ca1d51
75 Nac1f809726cc41fd984f172b0eb009aa schema:name dimensions_id
76 schema:value pub.1012359325
77 rdf:type schema:PropertyValue
78 Ndfbe1ba03b164c639f88e9ecc1ca1d51 rdf:first sg:person.016202177221.23
79 rdf:rest rdf:nil
80 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
81 schema:name Mathematical Sciences
82 rdf:type schema:DefinedTerm
83 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
84 schema:name Statistics
85 rdf:type schema:DefinedTerm
86 sg:journal.1125588 schema:issn 0885-6125
87 1573-0565
88 schema:name Machine Learning
89 rdf:type schema:Periodical
90 sg:person.016202177221.23 schema:affiliation https://www.grid.ac/institutes/grid.4836.9
91 schema:familyName Szepesvári
92 schema:givenName Csaba
93 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016202177221.23
94 rdf:type schema:Person
95 sg:person.07770302215.82 schema:affiliation https://www.grid.ac/institutes/grid.4836.9
96 schema:familyName Kocsis
97 schema:givenName Levente
98 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07770302215.82
99 rdf:type schema:Person
100 sg:pub.10.1007/978-1-4899-2696-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1019636471
101 https://doi.org/10.1007/978-1-4899-2696-8
102 rdf:type schema:CreativeWork
103 sg:pub.10.1007/bf00115009 schema:sameAs https://app.dimensions.ai/details/publication/pub.1028679285
104 https://doi.org/10.1007/bf00115009
105 rdf:type schema:CreativeWork
106 sg:pub.10.1007/bf00992696 schema:sameAs https://app.dimensions.ai/details/publication/pub.1001381236
107 https://doi.org/10.1007/bf00992696
108 rdf:type schema:CreativeWork
109 sg:pub.10.1007/bf00992697 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029271228
110 https://doi.org/10.1007/bf00992697
111 rdf:type schema:CreativeWork
112 sg:pub.10.1023/a:1007634325138 schema:sameAs https://app.dimensions.ai/details/publication/pub.1047698138
113 https://doi.org/10.1023/a:1007634325138
114 rdf:type schema:CreativeWork
115 https://app.dimensions.ai/details/publication/pub.1019636471 schema:CreativeWork
116 https://doi.org/10.1002/0471722138 schema:sameAs https://app.dimensions.ai/details/publication/pub.1109491948
117 rdf:type schema:CreativeWork
118 https://doi.org/10.1016/j.neucom.2004.11.016 schema:sameAs https://app.dimensions.ai/details/publication/pub.1023604540
119 rdf:type schema:CreativeWork
120 https://doi.org/10.1016/s0004-3702(01)00130-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1044471374
121 rdf:type schema:CreativeWork
122 https://doi.org/10.1016/s0020-0255(03)00045-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1049691245
123 rdf:type schema:CreativeWork
124 https://doi.org/10.1016/s0925-2312(01)00700-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052928518
125 rdf:type schema:CreativeWork
126 https://doi.org/10.1049/cp:19991170 schema:sameAs https://app.dimensions.ai/details/publication/pub.1098689874
127 rdf:type schema:CreativeWork
128 https://doi.org/10.1109/72.809083 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061219348
129 rdf:type schema:CreativeWork
130 https://doi.org/10.1109/9.119632 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061242810
131 rdf:type schema:CreativeWork
132 https://doi.org/10.1109/acc.1997.609490 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094523266
133 rdf:type schema:CreativeWork
134 https://doi.org/10.1109/icnn.1993.298623 schema:sameAs https://app.dimensions.ai/details/publication/pub.1086259851
135 rdf:type schema:CreativeWork
136 https://doi.org/10.1109/iconip.2002.1198180 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093470523
137 rdf:type schema:CreativeWork
138 https://doi.org/10.1109/ispa.2005.195385 schema:sameAs https://app.dimensions.ai/details/publication/pub.1093503311
139 rdf:type schema:CreativeWork
140 https://doi.org/10.1109/tac.2000.880982 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061474938
141 rdf:type schema:CreativeWork
142 https://doi.org/10.1109/tac.2003.815008 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061475376
143 rdf:type schema:CreativeWork
144 https://doi.org/10.1109/tnn.1998.712192 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061716400
145 rdf:type schema:CreativeWork
146 https://doi.org/10.1109/wsc.1998.744910 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095505241
147 rdf:type schema:CreativeWork
148 https://doi.org/10.1109/wsc.2002.1172897 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095737253
149 rdf:type schema:CreativeWork
150 https://doi.org/10.1137/s1052623495270723 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062883483
151 rdf:type schema:CreativeWork
152 https://doi.org/10.1214/aoms/1177698258 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064399247
153 rdf:type schema:CreativeWork
154 https://doi.org/10.1214/aoms/1177728659 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064401477
155 rdf:type schema:CreativeWork
156 https://doi.org/10.1214/aoms/1177729392 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064401671
157 rdf:type schema:CreativeWork
158 https://doi.org/10.1214/aoms/1177729586 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064401743
159 rdf:type schema:CreativeWork
160 https://doi.org/10.1214/aos/1059655913 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064406449
161 rdf:type schema:CreativeWork
162 https://doi.org/10.1214/aos/1176350965 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064409307
163 rdf:type schema:CreativeWork
164 https://doi.org/10.1287/mnsc.31.1.66 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064719947
165 rdf:type schema:CreativeWork
166 https://doi.org/10.1287/mnsc.38.6.884 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064720955
167 rdf:type schema:CreativeWork
168 https://doi.org/10.1287/mnsc.45.11.1570 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064721842
169 rdf:type schema:CreativeWork
170 https://doi.org/10.1613/jair.806 schema:sameAs https://app.dimensions.ai/details/publication/pub.1105579523
171 rdf:type schema:CreativeWork
172 https://www.grid.ac/institutes/grid.4836.9 schema:alternateName MTA Institute for Computer Science and Control
173 schema:name MTA SZTAKI, Kende u. 13–17, 1111, Budapest, Hungary
174 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...