Commit ed44e246 by Qi Guo

up

parent a5fa4be2
...@@ -47,7 +47,7 @@ ease-of-teaching~\cite{}. \note{The basic intuition is that high compositional s ...@@ -47,7 +47,7 @@ ease-of-teaching~\cite{}. \note{The basic intuition is that high compositional s
language cannot emerge without induction in existing multi-agent environment.} language cannot emerge without induction in existing multi-agent environment.}
Figure~\ref{fig:induction} reports the compositionality when training two agents in the widely-used Figure~\ref{fig:induction} reports the compositionality when training two agents in the widely-used
listener-speaker referential game, and it can be observed that \note{the compositionality listener-speaker referential game, and it can be observed that \note{the compositionality
of emerged symbolic language is extremely low}. of emerged symbolic language is extremely low without any induction}.
Though such unnatural inductions are useful, they prevent us from better understanding the mystery of 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. 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 symbolic language
...@@ -55,15 +55,19 @@ Yet, few works investigate the emergence of high compositional symbolic language ...@@ -55,15 +55,19 @@ Yet, few works investigate the emergence of high compositional symbolic language
In other words, it is never clear whether \emph{natural} In other words, it is never clear whether \emph{natural}
environment and agents are sufficient for achieving high compositionality. environment and agents are sufficient for achieving high compositionality.
In this work, we are the first to achieve high compositional In this paper, we are the first work to achieve high compositional
symbolic language without any handcrafted induction. Concretely, we first symbolic language without any deliberately handcrafted induction. Concretely,
thoroughly analyze the compositionality of emerged symbolic by thoroughly analyzing the compositionality of emerged symbolic
language by removing the \emph{deliberately handcrafted} language after removing the inductions in a listener-speaker referential game framework,
inductions. Figure~\ref{fig:comp} reports the compositionality when train two we observe that the agent capacity
agents in a listener-speaker referential game. It can be observed that \note{it plays a key role in improving the compositionality, and thus we can achieve
is challenging to achieve high compositionality without induction as a high compositional symbilic language in such a natural environment.
xxxxxx}. Moreover, we observe that the agent capacity plays a key role in
compositionality, see Figure xxx. %Figure~\ref{fig:comp} reports the compositionality when train two
%agents in a listener-speaker referential game. It can be observed that \note{it
% is challenging to achieve high compositionality without induction as
% xxxxxx}. Moreover, we observe that the agent capacity plays a key role in
%compositionality, see Figure xxx.
%focus on the natural emergence of high compositional symbolic language %focus on the natural emergence of high compositional symbolic language
%naturally without any handcrafted induction. %naturally without any handcrafted induction.
...@@ -75,16 +79,17 @@ compositionality, see Figure xxx. ...@@ -75,16 +79,17 @@ compositionality, see Figure xxx.
% xxxxxx}. Moreover, we observe that the agent capacity plays a key role in % xxxxxx}. Moreover, we observe that the agent capacity plays a key role in
%compositionality, see Figure xxx. %compositionality, see Figure xxx.
We reveal and characterize the quantitative relationship The quantitative relationship between the agent capacity and the compositionality
between the agent capacity and the compositionality of symbolic language both of symbolic language is revealed both theoretically and experimentally.
theoretically and experimentally.
%theoretically %theoretically
Regarding the theoretical analysis, we use the Regarding the theoretical analysis, we use the
Markov Series Channel (MSC)~\cite{} to model the language transmission process and a \note{Markov Series Channel (MSC)~\cite{} to model the language transmission process and a
novel mutual information-based metric to measure the compositionality quantitatively. novel mutual information-based metric to measure the compositionality quantitatively}.
%experimentally %experimentally
Regarding the experimental verification, it is conducted on a listener-speaker Regarding the experimental validation, two different dedicated experiments, i.e.,
referential game framework with eliminated unnatural inductions. \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 theoretical analysis and Both theoretical analysis and
experimental results lead to a counter-intuitive conclusion that lower agent 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 symbolic language with higher
...@@ -179,8 +184,9 @@ Both theoretical analysis and ...@@ -179,8 +184,9 @@ Both theoretical analysis and
In this paper, we made the following contributions: In this paper, we made the following contributions:
\begin{itemize}[topsep=0pt,itemsep=0cm] \begin{itemize}[topsep=0pt,itemsep=0cm]
\item We are the first to successfully achieve high compositional symbolic \item To our best knowledge, we are the first work to successfully achieve
language naturally, without any deliberately handcrafted inductions. high compositional symbolic
language naturally, without any deliberately handcrafted induction.
\item We thoroughly analyze the compositionality of emerged symbolic language \item We thoroughly analyze the compositionality of emerged symbolic language
after removing deliberately handcrafted inductions, and confirm that the agent after removing deliberately handcrafted inductions, and confirm that the agent
capacity acts as a key factor for compositionality. capacity acts as a key factor for compositionality.
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment