title={Emergence of Communication in an Interactive World with Consistent Speakers},
author={Bogin, Ben and Geva, Mor and Berant, Jonathan},
journal={arXiv},
pages={arXiv--1809},
year={2018}
}
@inproceedings{jaques2019social,
title={Social influence as intrinsic motivation for multi-agent deep reinforcement learning},
author={Jaques, Natasha and Lazaridou, Angeliki and Hughes, Edward and Gulcehre, Caglar and Ortega, Pedro and Strouse, DJ and Leibo, Joel Z and De Freitas, Nando},
booktitle={International Conference on Machine Learning},
pages={3040--3049},
year={2019},
organization={PMLR}
}
@article{mul2019mastering,
title={Mastering emergent language: learning to guide in simulated navigation},
author={Mul, Mathijs and Bouchacourt, Diane and Bruni, Elia},
journal={arXiv preprint arXiv:1908.05135},
year={2019}
}
@inproceedings{kharitonov2019egg,
title={EGG: a toolkit for research on Emergence of lanGuage in Games},
author={Kharitonov, Eugene and Chaabouni, Rahma and Bouchacourt, Diane and Baroni, Marco},
booktitle={Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations},
pages={55--60},
year={2019}
}
@article{labash2020perspective,
title={Perspective taking in deep reinforcement learning agents},
author={Labash, Aqeel and Aru, Jaan and Matiisen, Tambet and Tampuu, Ardi and Vicente, Raul},
journal={Frontiers in Computational Neuroscience},
\caption{The distribution of compositionality for 100 emerged symbolic
\caption{The distribution of compositionality for 100 emergent
languages without
any induction. It can be observed that high compositional symbolic language
any induction. It can be observed that high compositional language
seldom emerged (e.g., $<5\%$ for compositionality $>0.99$). Moreover, varying
the vocabulary size does not affect the compositionality notably.}
\label{fig:induction}
...
...
@@ -71,32 +71,34 @@ vocabulary can express almost infinite concepts.}
\end{tabular}
\end{table*}
Prior studies focus on achieving high compositional symbolic language
through \emph{deliberately handcrafted} inductions, e.g., small vocabulary
sizes~\cite{}, memoryless~\cite{}, additional rewards~\cite{}, constructed loss functions~\cite{}, and
ease-of-teaching~\cite{}. \note{Such optimization methodologies are driven by the challenges to generate high compositional symbolic without induction in an existing multi-agent environment.}
Prior studies focus on achieving high compositional language
through \emph{deliberately handcrafted} inductions, e.g., additional rewards~\cite{},
constructed loss functions~\cite{}, structural input data~\cite{},
memoryless~\cite{}, and ease-of-teaching~\cite{}.
\note{Such optimization methodologies are driven by the challenges to generate high compositional language without induction in an existing multi-agent environment.}
Figure~\ref{fig:induction} reports the compositionality when training two agents
in the widely-used listener-speaker referential game for emerging 100 symbolic
in the widely-used listener-speaker referential game~\cite{} for emerging 100
languages, and it can be observed that \note{the compositionality
of emerged symbolic language is extremely low without any induction. Moreover, varying
of emergent language is seldom high (e.g., $<5\%$ for compositionality $>0.99$)
without any induction. Moreover, varying
the vocabulary size does not affect the compositionality notably.}
Though such unnatural inductions are useful, they prevent us from better understanding the mystery of
the emergence of language and even intelligence among our pre-human ancestors.
Yet, few works investigate the emergence of high compositional symbolic language
Yet, few works investigate the emergence of high compositional language
\emph{naturally}, i.e., without handcrafted inductions.
In other words, it is never clear whether \emph{natural}
environment and agents are sufficient for achieving high compositionality.
This paper is the first one to achieve high compositional
symbolic language without any deliberately handcrafted induction. The key observation
language without any deliberately handcrafted induction. The key observation
is that the internal \emph{agent capacity} plays a crucial role in the
compositionality of symbolic language.
compositionality of emergent language.
%by thoroughly
%analyzing the compositionality after removing the inductions in
%the most widely-used listener-speaker referential game framework.
Concretely, the relationship between the agent capacity and the compositionality
of symbolic language is characterized, with a novel mutual information-based
Concretely, the relationship between the agent capacity and the compositionality
of emergent language is characterized, with a novel mutual information-based
metric for the compositionality.
%both theoretically and experimentally.
%theoretically
...
...
@@ -106,16 +108,16 @@ Regarding the theoretical analysis, we propose
a novel mutual information-based metric to measure the compositionality quantitatively.
%experimentally
Regarding the experimental validation, we exploit the relationship between agent
capacity and the compositionality of symbolic language that emerged
\emph{naturally}in our experiments.
capacity and the compositionality of \emph{naturally} emergent language
in our experiments.
%two different dedicated experiments, i.e., \note{XXX and XXX, are utilized for XXX}.
%Regarding the experimental validation, it is conducted on a listener-speaker
%referential game framework with eliminated unnatural inductions.
Both the theoretical analysis and experimental results lead to a counter-intuitive
conclusion that \emph{lower agent capacity facilitates the emergence of symbolic language
Both the theoretical analysis and experimental results lead to a counter-intuitive
conclusion that \emph{lower agent capacity facilitates the emergence of language
with higher compositionality}. \note{Therefore, by only reducing the agent capacity
in such a natural environment, we
can generate a higher compositional symbolic language with a higher probability.}
in such a natural environment, we
can generate a more compositional language with a higher probability.}
%Prior studies focus on investigating how to affect the
...
...
@@ -208,16 +210,16 @@ can generate a higher compositional symbolic language with a higher probability.
In this paper, we made the following contributions:
\begin{itemize}[topsep=0pt,itemsep=0cm]
\item To our best knowledge, we are the first work to successfully achieve
high compositional symbolic
high compositional emergent
language naturally, without any deliberately handcrafted induction.
\item We analyze the compositionality of emerged symbolic language
\item We analyze the compositionality of emergent language
after removing deliberately handcrafted inductions.
\item We propose a novel mutual information-based metric to measure the
compositionality quantitatively, which is more reasonable.
\item We experimentally exploited the relationship between agent
capacity. Both theoretical analysis and
experimental results lead to a counter-intuitive conclusion that lower agent
capacity facilitates the emergence of symbolic language with higher
capacity facilitates the emergence of language with higher
compositionality.
\end{itemize}
...
...
@@ -227,6 +229,6 @@ Section~\ref{sec:relatedwork} summarizes the related works.
Section~\ref{sec:thory}
introduces the experimental setup used in this study. Section~\ref{sec:mis}
describes our proposed novel mutual-information-based metric for measuring
the compositionality of symbolic language. Section~\ref{sec:exp} gives the
the compositionality of emergent language. Section~\ref{sec:exp} gives the
experimental results of the exploration for the relationship between agent
capacity and compositionality. Section~\ref{sec:con} concludes this paper.
