tex/gostyle.bib

   1 % This file was created with JabRef 2.6b2.
   2 % Encoding: UTF-8
   3
   4 @ELECTRONIC{GoR,
   5   author = {Ales Cieply and others},
   6   title = {EGF ratings system -- System description},
   7   owner = {pasky},
   8   timestamp = {2011.03.10},
   9   url = {http://www.europeangodatabase.eu/EGD/EGF_rating_system.php}
  10 }
  11
  12 @ELECTRONIC{ProGoR,
  13   author = {Ales Cieply and others},
  14   title = {Go Ratings of Professional and Strong Amateur Players},
  15   owner = {pasky},
  16   url = {http://www.goweb.cz/progor}
  17 }
  18
  19 @ELECTRONIC{KGSAnalytics,
  20   author = {Kazuhiro},
  21   title = {KGS Analytics},
  22   owner = {pasky},
  23   url = {http://kgs.gosquares.net/}
  24 }
  25
  26 @ELECTRONIC{Kombilo,
  27   author = {Ulrich G\"ortz},
  28   title = {Kombilo --- a go database program},
  29   owner = {pasky},
  30   url = {http://www.u-go.net/kombilo/}
  31 }
  32
  33 @ELECTRONIC{MoyoGo,
  34   author = {Frank de Groot},
  35   title = {Moyo Go Studio},
  36   owner = {pasky},
  37   url = {http://www.moyogo.com/}
  38 }
  39
  40 @ELECTRONIC{SGF,
  41   author = {Arno Hollosi},
  42   title = {SGF File Format},
  43   owner = {pasky},
  44   url = {http://www.red-bean.com/sgf/}
  45 }
  46
  47 @ELECTRONIC{GTP,
  48   author = {Gunnar Farneb\"{a}ck},
  49   title = {GTP --- Go Text Protocol},
  50   owner = {pasky},
  51   url = {http://www.lysator.liu.se/~gunnar/gtp/}
  52 }
  53
  54 @ELECTRONIC{RankComparison,
  55   author = {Sensei's Library},
  56   title = {Rank --- worldwide comparison},
  57   owner = {pasky},
  58   url = {http://senseis.xmp.net/?RankWorldwideComparison}
  59 }
  60
  61 @ELECTRONIC{GTL,
  62   author = {Jean-loup Gailly, Bill Hosken and others},
  63   title = {The Go Teaching Ladder},
  64   owner = {pasky},
  65   url = {http://gtl.xmp.net/}
  66 }
  67
  68 @ELECTRONIC{GoGoD,
  69   author = {T. Mark Hall, John Fairbairn},
  70   title = {Games of Go on Disk --- GoGoD Encyclopaedia and Database},
  71   owner = {pasky},
  72   url = {http://www.gogod.co.uk/}
  73 }
  74
  75 @ELECTRONIC{GoDiscThread,
  76   author = {GoDiscussions},
  77   title = {Pro Styles (discussion thread)},
  78   owner = {pasky},
  79   url = {http://www.godiscussions.com/forum/showthread.php?t=10980}
  80 }
  81
  82 @ELECTRONIC{Pachi,
  83   author = {Petr Baudi\v{s} and others},
  84   title = {Pachi --- Simple Go/Baduk/Weiqi Bot},
  85   owner = {pasky},
  86   url = {http://repo.or.cz/w/pachi.git}
  87 }
  88
  89 @ELECTRONIC{GoStyle,
  90   author = {Josef Moud\v{r}\'{i}k, Petr Baudi\v{s} and others},
  91   title = {GoStyle --- Determine playing style in the game of Go},
  92   owner = {pasky},
  93   url = {http://repo.or.cz/w/gostyle.git}
  94 }
  95
  96 @inproceedings{SpatPat,
  97         author = {Stern, David and Herbrich, Ralf and Graepel, Thore},
  98         title = {Bayesian pattern ranking for move prediction in the game of Go},
  99         booktitle = {ICML '06: Proceedings of the 23rd international conference on Machine learning},
 100         year = {2006},
 101         isbn = {1-59593-383-2},
 102         pages = {873--880},
 103         location = {Pittsburgh, Pennsylvania},
 104         doi = {http://doi.acm.org/10.1145/1143844.1143954},
 105         publisher = {ACM},
 106         address = {New York, NY, USA},
 107 }
 108
 109 @inproceedings{PatElo,
 110         title = { {C}omputing {E}lo {R}atings of {M}ove {P}atterns in the {G}ame of {G}o},
 111         author = {{C}oulom, {R}{\'e}mi},
 112         abstract = {{M}ove patterns are an essential method to incorporate domain knowledge into {G}o-playing programs. {T}his paper presents a new {B}ayesian technique for supervised learning of such patterns from game records, based on a generalization of {E}lo ratings. {E}ach sample move in the training data is considered as a victory of a team of pattern features. {E}lo ratings of individual pattern features are computed from these victories, and can be used in previously unseen positions to compute a probability distribution over legal moves. {I}n this approach, several pattern features may be combined, without an exponential cost in the number of features. {D}espite a very small number of training games (652), this algorithm outperforms most previous pattern-learning algorithms, both in terms of mean log-evidence (−2.69), and prediction rate (34.9%). {A} 19x19 {M}onte-{C}arlo program improved with these patterns reached the level of the strongest classical programs.},
 113         language = {{A}nglais},
 114         affiliation = {{SEQUEL} - {INRIA} {F}uturs - {INRIA} - {CNRS} : {UMR}8022 - {CNRS} : {UMR}8146 - {U}niversit{\'e} des {S}ciences et {T}echnologies de {L}ille - {L}ille {I} - {U}niversit{\'e} {C}harles de {G}aulle - {L}ille {III} - {E}cole {C}entrale de {L}ille },
 115         booktitle = {{C}omputer {G}ames {W}orkshop },
 116         address = {{A}msterdam {P}ays-{B}as },
 117         editor = {van den {H}erik, {H}. {J}aap and {M}ark {W}inands and {J}os {U}iterwijk and {M}aarten {S}chadd },
 118         audience = {non sp{\'e}cifi{\'e}e },
 119         year = {2007},
 120         URL = {http://hal.inria.fr/inria-00149859/en/},
 121         URL = {http://hal.inria.fr/inria-00149859/PDF/MMGoPatterns.pdf},
 122 }
 123         note = {{I}.: {C}omputing {M}ethodologies/{I}.2: {ARTIFICIAL} {INTELLIGENCE}/{I}.2.6: {L}earning, {I}.: {C}omputing {M}ethodologies/{I}.2: {ARTIFICIAL} {INTELLIGENCE}/{I}.2.8: {P}roblem {S}olving, {C}ontrol {M}ethods, and {S}earch/{I}.2.8.3: {G}raph and tree search strategies },
 124
 125 @ARTICLE{CoverHart1967,
 126   author = {Thomas M. Cover and Peter E. Hart},
 127   title = {Nearest neighbor pattern classification},
 128   journal = {IEEE Transactions on Information Theory},
 129   year = {1967},
 130   volume = {13},
 131   pages = {21-27},
 132   number = {1},
 133   owner = {hellboy},
 134   timestamp = {2010.03.10}
 135 }
 136
 137 @INPROCEEDINGS{GellySilver2008,
 138   author = {Gelly, Sylvain and Silver, David},
 139   title = {Achieving master level play in 9x9 computer go},
 140   booktitle = {AAAI'08: Proceedings of the 23rd national conference on Artificial
 141         intelligence},
 142   year = {2008},
 143   pages = {1537--1540},
 144   publisher = {AAAI Press},
 145   abstract = {The UCT algorithm uses Monte-Carlo simulation to estimate the value
 146         of states in a search tree from the current state. However, the first
 147         time a state is encountered, UCT has no knowledge, and is unable
 148         to generalise from previous experience. We describe two extensions
 149         that address these weaknesses. Our first algorithm, heuristic UCT,
 150         incorporates prior knowledge in the form of a value function. The
 151         value function can be learned offline, using a linear combination
 152         of a million binary features, with weights trained by temporal-difference
 153         learning. Our second algorithm, UCT-RAVE, forms a rapid online generalisation
 154         based on the value of moves. We applied our algorithms to the domain
 155         of 9 × 9 Computer Go, using the program MoGo. Using both heuristic
 156         UCT and RAVE, MoGo became the first program to achieve human master
 157         level in competitive play.},
 158   isbn = {978-1-57735-368-3},
 159   location = {Chicago, Illinois}
 160 }
 161
 162 @BOOK{Jolliffe1986,
 163   title = {Principal Component Analysis},
 164   publisher = {Springer, New York},
 165   year = {1986},
 166   author = {I.T. Jolliffe},
 167   owner = {hellboy}
 168 }
 169
 170 @BOOK{Elo,
 171   title = {The rating of chessplayers, past and present},
 172   publisher = {Arco, New York},
 173   year = {1978},
 174   author = {Arpad E. Elo},
 175   isbn = {0668047216},
 176   owner = {pasky}
 177 }
 178
 179 @ELECTRONIC{KohonenPy,
 180   author = {lmjohns3},
 181   title = {python-kohonen, {A} library of {Kohonen} maps},
 182   howpublished = {Released under {MIT} License},
 183   url = {http://code.google.com/p/python-kohonen/},
 184   owner = {hellboy},
 185   timestamp = {2010.03.10}
 186 }
 187
 188 @TECHREPORT{Nissen2003,
 189   author = {S. Nissen},
 190   title = {Implementation of a Fast Artificial Neural Network Library (fann)},
 191   institution = {Department of Computer Science University of Copenhagen (DIKU)},
 192   year = {2003},
 193   note = {http://fann.sf.net},
 194   owner = {hellboy},
 195   timestamp = {2010.03.10}
 196 }
 197
 198 @ARTICLE{Pearson,
 199   author = {J. L. Rodgers and W. A. Nicewander},
 200   title = {{Thirteen ways to look at the correlation coefficient}},
 201   journal = {The American Statistician},
 202   year = {1988},
 203   month = {Feb},
 204   volume = {42},
 205   number = {1},
 206   pages = {59--66},
 207   owner = {pasky}
 208 }
 209
 210 @ELECTRONIC{Python25,
 211   author = {{Python Software Foundation}},
 212   month = {September},
 213   year = {2006},
 214   title = {Python 2.5},
 215   url = {http://www.python.org/dev/peps/pep-0356/},
 216   owner = {hellboy},
 217   timestamp = {2009.04.29}
 218 }
 219
 220 @INPROCEEDINGS{Riedmiller1993,
 221   author = {Martin Riedmiller and Heinrich Braun},
 222   title = {{A Direct Adaptive Method for Faster Backpropagation Learning: The
 223         RPROP Algorithm}},
 224   booktitle = {IEEE International Conference on Neural Networks},
 225   year = {1993},
 226   pages = {586--591},
 227   owner = {hellboy},
 228   timestamp = {2010.03.07}
 229 }
 230
 231 @ARTICLE{MDP,
 232   author = {Zito Tiziano and Wilbert Niko and Wiskott Laurenz and Berkes Pietro},
 233   title = {Modular toolkit for {D}ata {P}rocessing ({MDP}): a {P}ython data
 234         processing framework},
 235   journal = {Frontiers in Neuroinformatics},
 236   year = {2008},
 237   volume = {2},
 238   abstract = {Modular toolkit for Data Processing (MDP) is a data processing framework
 239         written in Python. From the user&#39;s perspective, MDP is a collection
 240         of supervised and unsupervised learning algorithms and other data
 241         processing units that can be combined into data processing sequences
 242         and more complex feed-forward network architectures. Computations
 243         are performed efficiently in terms of speed and memory requirements.
 244         From the scientific developer&#39;s perspective, MDP is a modular
 245         framework, which can easily be expanded. The implementation of new
 246         algorithms is easy and intuitive. The new implemented units are then
 247         automatically integrated with the rest of the library. MDP has been
 248         written in the context of theoretical research in neuroscience, but
 249         it has been designed to be helpful in any context where trainable
 250         data processing algorithms are used. Its simplicity on the user&#39;s
 251         side, the variety of readily available algorithms, and the reusability
 252         of the implemented units make it also a useful educational tool.},
 253   doi = {10.3389/neuro.11/008.2008},
 254   issn = {ISSN 1662-5196},
 255   owner = {hellboy},
 256   timestamp = {2010.03.05}
 257 }
 258
 259 @comment{jabref-meta: selector_publisher:}
 260
 261 @comment{jabref-meta: selector_author:}
 262
 263 @comment{jabref-meta: selector_journal:}
 264
 265 @comment{jabref-meta: selector_keywords:}
 266