Commit 7c486653 by Zidong Du

~

parent 458544ad
......@@ -2,14 +2,6 @@
%\section{Agent Capacity vs. Compositionality}
%\label{ssec:exp}
\begin{figure}[t]
\centering \includegraphics[width=0.99\columnwidth]{fig/Figure6_Compostionality_of_symbolic_language.pdf}
\caption{Compositionality of symbolic language under different parameters
($[\mu-\sigma,\mu+\sigma]$, where $\mu$ is the mean value and $\sigma$ is
the standard deviation).}
\label{fig:exp1}
\end{figure}
\begin{figure}[t]
\centering \includegraphics[width=0.99\columnwidth]{fig/Figure7_The_ratio_of_high_compositional_language.pdf}
......@@ -29,6 +21,7 @@
\begin{table}[b]
\centering
\small
\caption{The Chi-square test between high-compositionality and agent capacity.}
\label{tab:exp10}
\begin{tabular}{cccc}
......
......@@ -11,6 +11,13 @@ MIS is the similarity between an identity matrix and the mutual information matr
\end{figure}
\begin{figure}[t]
\centering
\includegraphics[width=0.8\columnwidth]{fig/Figure5_An_emergent_language.pdf}
\caption{An emergent language that the unilateral metrics cannot measure its non-compositionality. Notice that given $s_1 = \mathrm{a}$, the listener can neither determine the shape nor the color without the knowledge about $s_0$.}
\label{fig:unilateral}
\end{figure}
Before giving the definition of MIS, we first model the agents in the referential games. As shown in Figure~\ref{fig:modeling}, the listener and speaker in the referential game are connected in tandem. The speaker agent can be regard as a channel, whose input is a concept $c = (c_0, c_1)$ and output is a symbol $s = (s_0, s_1)$. The listener agent can be regard as another channel, whose input is a symbol $s = (s_0, s_1)$ and output is a predict result $\hat{t} = (\hat{c}_0, \hat{c}_1)$. Since the output of the listener only depends on the symbol $s$, we can model the policy of the speaker agent and the listener agent by the probability distribution $P(s = (s_0, s_1) | t = (c_0, c_1))$ and $P(\hat{t} = (\hat{c}_0, \hat{c}_1) | s_0, s_1)$, respectively.
Now we can analyse the information of the concepts preserved in the transmission process given the symbol transmitted, i.e. the conditional mutual information $I\left(t,\hat{t}|s\right)$. Whenever a stable language emerged, the speaker and the listener consistently use a specific symbol $s$ to refer to a specific object $t$. Therefore we can safely say $I\left(t,\hat{t}|s\right) = I\left(t,\hat{t}|s_{t,\hat{t}}\right)$ where $s_{t,\hat{t}}=\max_s\left\{P\left(\hat{t}|s\right)P\left(s|t\right)\right\}$. This conditional mutual information can be obtained by Equation~\ref{eq:cmi}.
......@@ -33,15 +40,16 @@ R\left(c_0,s_0\right) & R\left(c_0,s_0\right)
\end{equation}
Each column of $M$ correspond to the semantic information carried by one symbol. In a perfectly compositional language, each symbol represents one specific concept exclusively. Therefore, the similarity between the columns of $M$ and a one-hot vector is align with the compositionality of the emergent language.
\begin{figure}[t]
\centering
\includegraphics[width=0.8\columnwidth]{fig/Figure5_An_emergent_language.pdf}
\caption{An emergent language that the unilateral metrics cannot measure its non-compositionality. Notice that given $s_1 = \mathrm{a}$, the listener can neither determine the shape nor the color without the knowledge about $s_0$.}
\label{fig:unilateral}
\centering \includegraphics[width=0.99\columnwidth]{fig/Figure6_Compostionality_of_symbolic_language.pdf}
\caption{Compositionality of symbolic language under different parameters
($[\mu-\sigma,\mu+\sigma]$, where $\mu$ is the mean value and $\sigma$ is
the standard deviation).}
\label{fig:exp1}
\end{figure}
Finally, we define \emph{raw mutual information similarity} ($\mathit{MIS}_0$)
as the average cosine similarity of $M$ columns and one-hot vectors, as
Equation~\ref{eq:mis2}. Furthermore, $\mathit{MIS}$ is the normalized mutual
......
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