Authorea

Joe Corneli some wording changes throughout about 9 years ago

Commit id: 1286a413ddcf53e696b0fb763e8cdf71cb64345f

deletions | additions

% \section{Connections} \label{sec:connections-to-formal-definition} The features of our model match matches and expands upon Merton's \citeyear{merton1948bearing}earlier description quite well: of the ``serendipity pattern.'' $T$ is the an unexpected observation; $T^\star$ highlights its interesting or anomalous features and casts it recasts them as ``strategic data''; and and, finally, the result $R$ may include updates to $p$ or $p^{\prime}$ that inform further phases of research.The connection to the key condition and components of serendipity introduced in our literature survey are as follows: % The \textbf{focus shift} corresponds to the identification of $T^\star$, which is common to both the discovery and the invention phase. If the process operates in an ``online'' manner, $T^\star$ may be an evolving vector of interesting possibilities. % The \textbf{prepared mind} corresponds to the prior training $p$ and $p^{\prime}$ in our diagram. % The \textbf{serendipity trigger} is denoted by $T$ in our diagram. % The \textbf{bridge} is comprised of the actions based on $p^{\prime}$ that are taken on $T^\star$ leading to the \textbf{result} $R$. Although they do not directly figure in our definition, the supportive dimensions and factors can be interpreted using this schematic, as follows: % schematic to flesh out the description of serendipity in working systems. From the point of view of this model, the system under consideration, $T$ is indeterminate. Furthermore, one must assume that relatively few of triggers $T^\star$ that are identified as interesting actually lead to useful results; in other words, the process is fallible and \textbf{chance} is likely to play a role. % The prior training $p$ causes interesting features to be extracted, even if they are not necessarily useful; $p^{\prime}$ asks how these features \emph{might} be useful. These routines suggest the relevance of a computational model of \textbf{curiousity}.One existing algorithmic approach is developed by \citeA{schmidhuber2007simple}. % Rather than a simple look-up rule, $p^{\prime}$ involves creating new knowledge. A simple example is found in clustering systems, which generate new categories on the fly. A more complicated example, necessary in the case of updating $p$ or $p$ $p^{\prime}$, is automatic programming. There is ample room for \textbf{sagacity} in this affair. % Judging Judgment of the \textbf{value} of the result $R$ may be carried out ``locally'' (as an embedded part of the process of invention of $R$) or ``globally'' (i.e.~as an external process). % As noted above, noted, $T$ (and $T^\star$) appears within a stream of data with indeterminacy. There is a further an additional feedback loop, insofar as products $R$ influence the future state and behaviour of the system. Thus, the model system exists in a \textbf{dynamic world}. % Our model separates the ``context of discovery'', involving prior preparations $p$, from the ``context of invention'' involving prior preparations $p^{\prime}$. Both of these these, and the data they deal with, may be subdivided further into \textbf{multiple contexts}. % Both And correspondingly, since both $T$ and $T^\star$ may be multiple, causing an individual process to fork into complex, they may be processed using multiple sub-processes dealing that deal with \textbf{multiple tasks} that involve using different skills sets. % The process as a whole may be multiplied out among across different communicating investigators, so that the final result bears the mark of \textbf{multiple influences}.

The connection to the key condition and components of serendipity introduced in our literature survey are as follows: % The \textbf{serendipity trigger} is denoted by $T$. % The \textbf{focus shift} takes place with the identification of $T^\star$, which is common to both the discovery and the invention phase. If the process operates in an ``online'' manner, $T^\star$ may be an evolving vector of interesting possibilities. % The \textbf{prepared mind} corresponds to the prior training $p$ and $p^{\prime}$ in our diagram. % % The \textbf{bridge} is comprised of the actions based on $p^{\prime}$ that are taken on $T^\star$ leading to the \textbf{result} $R$, which is ultimately given a positive evaluation. %% Here, $T$ is the trigger and $p$ denotes those preparations that afford the %% classification $T^\star$, indicating $T$ to be of interest, while %% $p^{\prime}$ denotes the preparations that facilitate the creation of a %% bridge to a result $R$, which is ultimately given a positive %% evaluation.

as their Highness travelled, they were always making discoveries, by accidents \& sagacity, of things which they were not in quest of}[.]''~\cite[p. 633]{van1994anatomy} \end{quote} The term became more widely known in the 1940s through studies of serendipity as a factor in scientific discovery, surveyed by Robert Merton and Elinor Barber \citeyear{merton} intheir 1957 analyis of ``The Travels and Adventures of Serendipity, A Study in Historical Semantics and the Sociology of Sciences''. Merton \citeyear{merton1948bearing} \cite[pp. 195--196]{merton} describes a generalised ``serendipity pattern'' and Barber define the term as follows: its constituent parts: \begin{quote} \emph{``The ``\emph{The serendipity pattern refers to the fairly common experience of observing an unanticipated, anomalous and strategic \emph{unanticipated}, \emph{anomalous} \emph{and strategic} datum which becomes the occasion for developing a new theory or for extending an existing theory.''} \cite[p. 635]{van1994anatomy} theory.}''~\cite[p. 506]{merton1948bearing} (original emphasis) %% The datum [that exerts a pressure for initiating theory] is, first of all, unanticipated. A research directed toward the test of one hypothesis yields a fortuitous by-product, an unexpected observation which bears upon theories not in question when the research was begun. %% Secondly, the observation is anomalous, surprising, either because it seems inconsistent with prevailing theory or with other established facts. In either case, the seeming inconsistency provokes curiosity; it stimulates the investigator to "make sense of the datum," to fit it into a broader frame of knowledge.... %% And thirdly, in noting that the unexpected fact must be "strategic," i. e., that it must permit of implications which bear upon generalized theory, we are, of course, referring rather to what the observer brings to the datum than to the datum itself. For it obviously requires a theoretically sensitized observer to detect the universal in the particular. \end{quote} In 1986, Philippe Qu\'eau described serendipity as ``the art of finding what we are not looking for by looking for what we are not finding'' \cite{eloge-de-la-simulation}, as quoted in

\subsection{Future Work} \label{sec:futurework} \label{sec:hatching} Within the context of the ongoing COINVENT project \cite{coinvent14}, we are interested in using computationalblending theory blending to realise certain aspects of this model in a stand-alone architecture. % It will be useful to consider how we can take both the \emph{discovery step}, which combines a serendipity trigger $T$, and prior preparation $p$ and produces $p$, to produce a classification $T^{\star}$ -- and the \emph{invention step}, which combines the classified trigger $T^{\star}$, and preparations $p^{\prime}$, and produces a novel result $R$ -- to be \emph{blends} in the sense of Joseph Goguen \citeyear{goguen1999introduction}. The epistemological framework of discovery gives some important clues about how to compute a common base between $T$ and $p$. $p$, a key step for blending. Although $T$ was previously uninteresting, it will have attributes or attribute-types that match the patterns recognised by $p$ (e.g. van Andel's \citeyear{van1994anatomy} ``\emph{One surprising observation}''). % In the invention step, reasoning, experimentation, social interaction strategies rely on $p^{\prime}$, which might draw on patterns like van Andel's \emph{Successful error} in order to extract a fruitful pinpoint the seeds of useful result from inside $T^{\star}$. Here, an One important guidepost for implementation is the theory-building orientation that says thatmany outcomes will should result in new patterns of behaviour that the system can draw on in subsequent interactions. What is particularly needed is an approach to encoding patterns of serendipity and methods for pattern discovery in a computationally accessible manner. Here we are drawn to the approach taken by the design pattern community, \emph{design pattern} community \cite{alexander1999origins}, although we recognize that we would be using design patterns in a very different way from the standard: rather nonstandard way: \begin{itemize} \item[(1)] We want to encode our ``design patterns'' design patterns directly in runnable programs, not just give them to programmers as heuristic guidance. \item[(2)] We want the automated programming system to generate new patterns, not just apply or adapt old ones. \end{itemize}

possible: Whereas most Men study to render it intricate and obscure.}~\cite[p. 54]{zadig} \end{quote} %% Men were much to blame for casting Reflections on the Conduct of %% the Whole, upon the bare Inspection of the minutest Part % In the Ruins, he will find an immense Treasure, that % will enable him, from henceforth, to exert his Beneficence, and render % his Virtues more and more conspicuous. %% Topick they entred upon was the Passions. Alas! said Zadig, how fatal %% in their Consequences! However, said the Hermit, they are the Winds %% that swell the Sail of the Vessel. Sometimes, ’tis true, they overset %% it; but there is no such Thing as sailing without them. Phlegm, %% indeed, makes Men peevish and sick; but then there is no living %% without it. Tho’ [215]every Thing here below is dangerous, yet All are %% necessary. In the next Place, their Discourse turn’d on sensual %% Pleasures; and the Hermit demonstrated, that they were the Gifts of %% Heaven; for, said he, Man cannot bestow either Sensations or Ideas on %% himself; he receives them all; his Pain and Pleasure, as well as his %% Being, proceed from a superior Cause. %% Mankind in general, said the Angel Jesrad, judge of the Whole, by only %% viewing the hither Link of the Chain.

\subsection{Recommendations} \label{sec:recommendations} % Deleuze writes: ``True freedom lies in the power to decide, to constitute problems themselves'' \cite[p. 15]{deleuze1991bergsonism}; and, elsewhere, rephrasing this sentiment in a social way:

be developed in heterogeneity rather than propose gestures for us to reproduce.}''~\cite[p. 26]{deleuze1994difference} \end{quote} Dewey emphasisedthe necessary feature of a child's training is that must dealonly with objects which ``arise out of their interests and their own problems'' \cite[p. 73]{dewey-by-mead}. Von Foerster advocated an approach to a form of cybernetics in which ``the observer who enters the system shall be allowed to stipulate his own purpose'' \cite[p. 286]{von2003essays}. % Whitehead is similar too.

The thought experiment presented in Section \ref{sec:ww} illustrated the relationship between problem creation and serendipity. Looking for the connections that make raw data into ``strategic data'' is a core pattern for of problem creation (see Section \ref{sec:hatching}). creation. This is an appropriate theme for researchers in computational creativity to grapple with. In \cite{stakeholder-groups-bookchapter}, we outlined a general programme for computational creativity, and examined perceptions of

recalling Turing's proposal that computers should ``be able to converse with each other to sharpen their wits'' \cite{turing-intelligent}. Other fields, including computer Chess, Go, and argumentation have achieved this, such standards, and to good effect. The Writers Workshop described in Section \ref{sec:ww} is an example of one such social model, but more fundamentally, it is an example of

\subsection{Related work} \label{sec:related} In a key theoretical work on serendipity, Merton \citeyear{merton1948bearing} \cite[pp. 195--196]{merton} refers to a generalised ``serendipity pattern'' and its constituent parts: \begin{quote} \emph{The serendipity pattern refers to the fairly common experience of observing an \emph{unanticipated}, \emph{anomalous} \emph{and strategic} datum which becomes the occasion for developing a new theory or for extending an existing theory.}~\cite[p. 506]{merton1948bearing} (original emphasis) %% The datum [that exerts a pressure for initiating theory] is, first of all, unanticipated. A research directed toward the test of one hypothesis yields a fortuitous by-product, an unexpected observation which bears upon theories not in question when the research was begun. %% Secondly, the observation is anomalous, surprising, either because it seems inconsistent with prevailing theory or with other established facts. In either case, the seeming inconsistency provokes curiosity; it stimulates the investigator to "make sense of the datum," to fit it into a broader frame of knowledge.... %% And thirdly, in noting that the unexpected fact must be "strategic," i. e., that it must permit of implications which bear upon generalized theory, we are, of course, referring rather to what the observer brings to the datum than to the datum itself. For it obviously requires a theoretically sensitized observer to detect the universal in the particular. \end{quote} As we will explain in the following section, the 13 components mentioned above can contribute to unpacking this definition. Paul Andr{\'e} et al.~\citeyear{andre2009discovery} look at serendipity from a design point of view. perspective. These authors also propose a two-part model, in which what we have called \emph{discovery} above exposes the unexpected, while \emph{invention} is the responsibility another subsystem that finds applications. According to Andr\'e et

reminiscent of the operating strategy of {\sf SHRDLU} \cite{winograd1972understanding}. Importantly, one of the most relevant preparations would be prior participation in Workshop dialogues. A system with prior experience in the Workshop may have a catalogue of outstanding unresolved, or partially resolved, problems (denoted ``X'' in the schematic above). Embodied in code, they may drive comments, questions, and other behaviour -- and they may be answered in unexpected ways. \paragraph{Thought Experiment: Serendipity triggers.} Although the poem is under the control of the initial generative

related to poetry (e.g.~definitions of words, valence of sentiments, metre, repetition, density, etc.) and code (e.g.~the data, functions, and macros in which the poetic concepts and workshop protocols are embodied). Previous Some notable previous experiments with concept invention have been fraught with questions about autonomy \cite{ritchie1984case,lenat1984and}. \textbf{[Some comment about HR here?]} One cognitively inspired hypothesis is that the formation of new concepts is closely related to formation of sensory experiences \cite{milan2013kiki}. If the workshop participants have the capacity to identify the distinctive

genetic algorithm approach could be used assemble a battery of existing low-level tools that can approximate the effect. Relatedly, a compression process could seek to produce a given complex poetic effect with a maximally-succinct algorithm. algorithm \cite{schmidhuber2007simple}. The key point is that feedback on the poem -- simply describing what's in the poem from several different points of view -- can be used to