Commit c499806f by Zidong Du
parents a63f1ece b20789e1
......@@ -178,18 +178,18 @@
The emergence of symbolic languages with high compositionality has
attracted extensive attention from a broad range of communities. Existing
studies achieve high compositionality through \emph{deliberately handcrafted}
inductions (e.g., small vocabulary sizes, carefully constructed distractors,
and ease-of-teaching) in multi-agent learning, which are unnatural.
inductions (e.g., small vocabulary sizes, addtional rewards, constructed
loss functions and ease-of-teaching) in multi-agent learning, which are unnatural.
Yet, few studies investigate the emergence of symbolic language with high
compositionality \emph{naturally}, i.e., without deliberately handcrafted
inductions.
In this paper, we are the first to successfully achieve high compositional symbolic
language in a \emph{natural} manner without handcrafted inductions.
In this paper, \note{we are the first to successfully achieve high compositional
symbolic language} in a \emph{natural} manner without handcrafted inductions.
Initially, by investigating the compositionality of emerged symbolic
language after removing the \emph{deliberately handcrafted}
inductions, we observe that the agent capacity plays a key role in
compositionality. We further reveal and characterize the quantitative relationship
compositionality. We further reveal and characterize the \note{quantitative relationship}
between the agent capacity and the compositionality of symbolic language, with
a novel mutual information-based metric for the compositionality.
% both theoretically and experimentally.
......
......@@ -33,19 +33,26 @@ agent itself. \rmk{this should be largely emphasized.}
%measure
To measure the compositionality of emerged symbolic language, many metrics are
proposed~\cite{}.
Widely accepted metrics can be classified into two categories, measuring
positive signaling~\cite{} and measuring positive listening~\cite{}. The former
metrics measure the relationship between spoken symbols and received concepts
\rmk{not clear}, from the perspective of \emph{speakers}.
For example,.
The latter metrics measure the relationship between received symbols and
predicted concepts \rmk{not clear}, from the perspective of \emph{listeners}.
For example,.
However, these metrics are not appropriate, for they only measure
compositionality of symbolic language in \emph{unilateral} role\rmk{not sure},
either speakers or listeners. They can not measure the degree of \emph{bilateral}
understanding between speakers and listeners, i.e., the concept-symbol mapping
consistency between speakers and listeners.
%Widely accepted metrics can be classified into two categories, measuring
%positive signaling~\cite{} and measuring positive listening~\cite{}. The former
%metrics measure the relationship between spoken symbols and received concepts
%\rmk{not clear}, from the perspective of \emph{speakers}.
%For example,.
%The latter metrics measure the relationship between received symbols and
%predicted concepts \rmk{not clear}, from the perspective of \emph{listeners}.
%For example,.
%However, these metrics are not appropriate, for they only measure
%compositionality of symbolic language in \emph{unilateral} role\rmk{not sure},
%either speakers or listeners. They can not measure the degree of \emph{bilateral}
%understanding between speakers and listeners, i.e., the concept-symbol mapping
%consistency between speakers and listeners.
At the initial stage, many researches only analyzed the language compositionality qualitatively.
For example, ~\citet{choi2018compositional} printed the agent messages with the letter `abcd' at some training round, and directly analyzed the compositionality on these messages.
~\citet{kottur-etal-2017-natural} introduced the dialog tree to show the evolution of language compositionality during the trianing process.
Latter, some quantitative metrics are explored.
The topographic similarity\cite{lazaridou2018emergence} is introduced to measure the distances between all the possible pairs of meanings and the corresponding pairs of signals.
\citet{chaabouni2020compositionality} proposed the positional disentanglement and the bag-of-symbols disentanglement. The positional disentanglement measures whether symbols in specific postion clearly relate to the specific attribute of the input object. The bag-of-symbols measure the permutation-invariant characteristic of a language.
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