Commit b5d699dd by Qi Guo

up

parent 8a9bd6bd
......@@ -43,7 +43,7 @@ More recently, measuring the compositionality \note{xxxxx}.}
Prior studies focus on achieving high compositional symbolic language
through \emph{deliberately handcrafted} inductions, e.g., small vocabulary
sizes~\cite{}, memoryless~\cite{}, carefully constructed rewards~\cite{}, and
ease-of-teaching~\cite{}. \note{The basic intuition is that high compositional symbolic
ease-of-teaching~\cite{}. \note{The possible intuition is that high compositional symbolic
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
listener-speaker referential game, and it can be observed that \note{the compositionality
......@@ -56,31 +56,13 @@ In other words, it is never clear whether \emph{natural}
environment and agents are sufficient for achieving high compositionality.
In this paper, we are the first work to achieve high compositional
symbolic language without any deliberately handcrafted induction. Concretely,
by thoroughly analyzing the compositionality of emerged symbolic
language after removing the inductions in the most widely-used listener-speaker
referential game framework, we observe that the agent capacity
plays a key role in improving the compositionality, and thus we can achieve
a high compositional symbolic language in such a natural environment.
%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
%naturally without any handcrafted induction.
%Initially, we thoroughly analyze the compositionality of emerged symbolic
%language after removing the \emph{deliberately handcrafted}
%inductions. 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.
The quantitative relationship between the agent capacity and the compositionality
of symbolic language is revealed both theoretically and experimentally.
symbolic 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,
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 both theoretically and experimentally.
%theoretically
Regarding the theoretical analysis, we use the
\note{Markov Series Channel (MSC)~\cite{} to model the language transmission process and a
......@@ -90,10 +72,11 @@ Regarding the experimental validation, 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 theoretical analysis and
experimental results lead to a counter-intuitive conclusion that lower agent
capacity facilitates the emergence of symbolic language with higher
compositionality.
Both the theoretical analysis and experimental results lead to a counter-intuitive
conclusion that \emph{lower agent capacity facilitates the emergence of symbolic language
with higher compositionality}. \note{Therefore, by only reducing the agent capacity, we
can generate a higher compositional symbolic language in such a natural environment
with a higher probability.}
%Prior studies focus on investigating how to affect the
......
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