more figures

Section06-Neural-Machine-Translation/section06-test.tex

@@ -145,6 +145,11 @@
     \item 整个句子编码到一个向量里可能会有信息丢失
     \item 缺少源语-目标语词之间词语的对应。某种意义上讲，一个目标语单词的生成无法区分不同源语单词的贡献
     \end{itemize}
+\item 但是，翻译是具有很强的局部性的，有些词之间会有更紧密的关系，这种关系可以在建模中考虑
+    \begin{itemize}
+    \item 源语词和目标语词的对应并不是均匀的，甚至非常稀疏
+    \item 比如，一些短语的生成仅依赖于源文中的少数词
+    \end{itemize}
 \end{itemize}
 
 \begin{center}
@@ -173,19 +178,27 @@
 \node [] (wt5) at (t5) {\scriptsize{long}};
 \node [] (wt6) at ([xshift=1em]t6) {\scriptsize{sentence}};
 
-\node [anchor=south west,fill=red!30,minimum width=1.6in,minimum height=1.5em] (encoder) at ([yshift=1em]ws1.north west) {\footnotesize{Encoder}};
+\node [anchor=south west,fill=red!30,minimum width=1.6in,minimum height=1.5em] (encoder) at ([yshift=1.0em]ws1.north west) {\footnotesize{Encoder}};
 \node [anchor=west,fill=blue!30,minimum width=1.9in,minimum height=1.5em] (decoder) at ([xshift=4.5em]encoder.east) {\footnotesize{Decoder}};
 \node [anchor=west,fill=green!30,minimum height=1.5em] (representation) at ([xshift=1em]encoder.east) {\footnotesize{表示}};
 \draw [->,thick] ([xshift=1pt]encoder.east)--([xshift=-1pt]representation.west);
 \draw [->,thick] ([xshift=1pt]representation.east)--([xshift=-1pt]decoder.west);
 
 \foreach \x in {1,2,...,6}
-    \draw[->] (s\x.north) -- ([yshift=0.5em]s\x.north);
+    \draw[->] ([yshift=0.1em]s\x.north) -- ([yshift=1.2em]s\x.north);
+    
+\foreach \x in {1,2,...,5}
+    \draw[<-] ([yshift=0.1em]t\x.north) -- ([yshift=1.2em]t\x.north);
+
+ \draw[<-] ([yshift=0.1em,xshift=1em]t6.north) -- ([yshift=1.2em,xshift=1em]t6.north);
 
 
 \end{scope}
 \end{tikzpicture}
 \end{center}
+\vspace{-1.5em}
+
+
 
 \end{frame}
 

Section06-Neural-Machine-Translation/section06.tex

@@ -882,7 +882,14 @@ NLP问题的隐含结构假设 & 无隐含结构假设，端到端学习 \\
 %%%------------------------------------------------------------------------------------------------------------
 %%% 一些变种
 \begin{frame}{改进 - 双向模型}
-%%% 图
+\begin{itemize}
+\item 自左向右的模型只考虑了左侧的上下文，因此可以用自右向左的模型对右侧上下文建模
+	\begin{itemize}
+	\item 最终将两个模型融合同时送给编码端
+	\end{itemize}
+\end{itemize}
+
+\vspace{-0.5em}
 \begin{center}
     \begin{tikzpicture}
         \setlength{\base}{0.9cm}
@@ -918,7 +925,7 @@ NLP问题的隐含结构假设 & 无隐含结构假设，端到端学习 \\
 
             % RNN Decoder
             \foreach \x in {1,2,...,10}
-                \node[rnnnode,minimum height=0.5\base,fill=green!30!white,anchor=south] (demb\x) at ([yshift=2\base]enc\x.north) {};
+                \node[rnnnode,minimum height=0.5\base,fill=green!30!white,anchor=south] (demb\x) at ([yshift=1.5\base]enc\x.north) {};
             \foreach \x in {1,2,...,10}
                 \node[rnnnode,fill=blue!30!white,anchor=south] (dec\x) at ([yshift=0.5\base]demb\x.north) {};
             \foreach \x in {1,2,...,10}
@@ -1007,7 +1014,7 @@ NLP问题的隐含结构假设 & 无隐含结构假设，端到端学习 \\
                 \draw[-latex'] (backenc\x.west) to (backenc\y.east);
             }
             \coordinate (bridge) at ([yshift=-1.2\base]demb2);
-            \draw[-latex'] (enc10.north) .. controls +(north:\base) and +(east:1.5\base) .. (bridge) .. controls +(west:2.5\base) and +(west:0.6\base) .. (dec1.west);
+            \draw[-latex'] (enc10.north) .. controls +(north:0.7\base) and +(east:1.5\base) .. (bridge) .. controls +(west:2.5\base) and +(west:0.6\base) .. (dec1.west);
             \draw[-latex'] (backenc1) to [out=180,in=180] (dec1.west);
 
             % Backward RNN