writing: 385562f7e51bef00ace4a3ca2e8182665e7b95af

     1: \iffalse
     2: \documentclass[]{article}
     3: 
     4: \usepackage{cite}
     5: \usepackage{rotating}
     6: \usepackage{hyperref}
     7: \usepackage{graphicx}
     8: \usepackage{mathtools}
     9: \usepackage{amssymb}
    10: \usepackage[T1]{fontenc}
    11: % \usepackage{enumitem}
    12: \usepackage{paralist}
    13: \usepackage{csquotes}
    14: \usepackage[affil-it]{authblk}
    15: \usepackage{listings}
    16: \usepackage{fixltx2e}
    17: \usepackage[numbers]{natbib}
    18: \usepackage{mathpartir}
    19: \usepackage{mmm}
    20: 
    21: % Treat paragraph as subsubsubsection
    22: \usepackage{titlesec}
    23: \setcounter{secnumdepth}{4}
    24: 
    25: \DeclareMathOperator{\md5}{MD5}
    26: 
    27: \newcommand{\blank}{\cdot}
    28: 
    29: \fi
    30: 
    31: \documentclass[runningheads,a4paper]{llncs}
    32: 
    33: \usepackage{amsmath}
    34: \usepackage{amssymb}
    35: \setcounter{tocdepth}{3}
    36: \usepackage{graphicx}
    37: \usepackage{minted}
    38: \usepackage[caption=false]{subfig}
    39: 
    40: \usepackage{tikz-qtree}
    41: \usepackage{tikz}
    42: \usetikzlibrary{shapes,arrows,decorations.pathreplacing,calc}
    43: 
    44: \usepackage{algorithm}
    45: \usepackage[noend]{algpseudocode}
    46: 
    47: \usepackage{url}
    48: \urldef{\mailsa}\path|cmwarburton@dundee.ac.uk, katya@computing.dundee.ac.uk|
    49: \newcommand{\keywords}[1]{\par\addvspace\baselineskip
    50: \noindent\keywordname\enspace\ignorespaces#1}
    51: 
    52: % Abbreviations
    53: \newcommand{\qcheck}{\textsc{QuickCheck}}
    54: \newcommand{\qspec}{\textsc{QuickSpec}}
    55: \newcommand{\hspec}{\textsc{HipSpec}}
    56: \newcommand{\equal}{=}
    57: \newcommand{\fc}{System $F_C$}
    58: 
    59: % Semantic markup
    60: \providecommand{\coq}[1]{\texttt{#1}}
    61: \newenvironment{haskell}{\ttfamily}{\par}
    62: \newenvironment{haskellblock}{\begin{minted}{haskell}}{\end{minted}}
    63: \newenvironment{coqblock}{\ttfamily}{\par}
    64: \newcommand{\hs}[1]{\texttt{#1}}
    65: \newcommand*\vect[1]{\mathbf{#1}}
    66: \newcommand*\mean[1]{\overline{#1}}
    67: \newcommand{\argmin}{\operatornamewithlimits{argmin}}
    68: \newcommand{\feature}[1]{\phi(#1)}
    69: \newcommand{\id}[1]{\texttt{"#1"}}
    70: \newcommand{\vlocal}[1]{\CVar\ (\CLocal\ \id{#1})}
    71: \newcommand{\vglobal}[1]{\CVar\ (\CGlobal\ \id{#1})}
    72: \newcommand{\cat}{\mbox{\ensuremath{+\!\!+\,}}}
    73: 
    74: % Shorthand for our diagrams
    75: \newcommand{\CVar}{\texttt{Var}}
    76: \newcommand{\CLit}{\texttt{Lit}}
    77: \newcommand{\CApp}{\texttt{App}}
    78: \newcommand{\CLam}{\texttt{Lam}}
    79: \newcommand{\CLet}{\texttt{Let}}
    80: \newcommand{\CCase}{\texttt{Case}}
    81: \newcommand{\CType}{\texttt{Type}}
    82: \newcommand{\CLocal}{\texttt{Local}}
    83: \newcommand{\CGlobal}{\texttt{Global}}
    84: \newcommand{\CConstructor}{\texttt{Constructor}}
    85: \newcommand{\CLitNum}{\texttt{LitNum}}
    86: \newcommand{\CLitStr}{\texttt{LitStr}}
    87: \newcommand{\CAlt}{\texttt{Alt}}
    88: \newcommand{\CDataAlt}{\texttt{DataAlt}}
    89: \newcommand{\CLitAlt}{\texttt{LitAlt}}
    90: \newcommand{\CDefault}{\texttt{Default}}
    91: \newcommand{\CNonRec}{\texttt{NonRec}}
    92: \newcommand{\CRec}{\texttt{Rec}}
    93: \newcommand{\CBind}{\texttt{Bind}}
    94: 
    95: \begin{document}
    96: 
    97: \mainmatter  % start of an individual contribution
    98: 
    99: % first the title is needed
   100: \title{Scaling Theory Exploration Through Bucketing}
   101: 
   102: % a short form should be given in case it is too long for the running head
   103: \titlerunning{Improving Haskell Theory Exploration}
   104: 
   105: % the name(s) of the author(s) follow(s) next
   106: %
   107: % NB: Chinese authors should write their first names(s) in front of
   108: % their surnames. This ensures that the names appear correctly in
   109: % the running heads and the author index.
   110: %
   111: \author{Chris Warburton%
   112: \and Ekaterina Komendantskaya}
   113: %
   114: \authorrunning{Improving Haskell Theory Exploration}
   115: % (feature abused for this document to repeat the title also on left hand pages)
   116: 
   117: % the affiliations are given next; don't give your e-mail address
   118: % unless you accept that it will be published
   119: \institute{University of Dundee,\\
   120: \mailsa\\
   121: \url{http://tocai.computing.dundee.ac.uk}}
   122: 
   123: %
   124: % NB: a more complex sample for affiliations and the mapping to the
   125: % corresponding authors can be found in the file "llncs.dem"
   126: % (search for the string "\mainmatter" where a contribution starts).
   127: % "llncs.dem" accompanies the document class "llncs.cls".
   128: %
   129: 
   130: %\toctitle{Lecture Notes in Computer Science}
   131: \tocauthor{Scaling Theory Exploration Through Bucketing}
   132: \maketitle
   133: 
   134: \begin{abstract}
   135:   Mathematical Theory Exploration can discover novel, interesting properties of
   136:   formal definitions (such as a mathematical theory, scientific model or
   137:   computer program). In principle this is a fully automated process, but in
   138:   practice requires careful selection of definitions to avoid infeasible running
   139:   times when more than a few dozen definitions are provided.
   140: 
   141:   We call this cherry-picking the \emph{bucketing problem}, and demonstrate its
   142:   necessity empirically by exploring definitions sampled from a theorem-proving
   143:   corpus; finding that problematic definitions severely increase the variance in
   144:   running time with size, despite the median remaining low. We measure the
   145:   effect of bucketing on the number of desirable theorems that remain possible
   146:   to discover, finding a large gap between best- and worst-case selections,
   147:   indicating the potential for smart bucketing strategies to avoid impacting
   148:   exploration potential too severely. We compare bucketing to \emph{clustering}
   149:   (grouping based on similarity), and find solutions to the latter perform no
   150:   better than random; indicating the difficulty of this open problem.
   151: \end{abstract}
   152: 
   153: \input{intro.tex}
   154: \input{background.tex}
   155: \input{contributions.tex}
   156: \input{sizes.tex}
   157: \input{implementation.tex}
   158: \input{evaluation.tex}
   159: \input{related.tex}
   160: \input{conclusion.tex}
   161: 
   162: \bibliographystyle{plain}
   163: \bibliography{Bibtex}
   164: 
   165: \input{appendices.tex}
   166: 
   167: \end{document}

Generated by git2html.