TP L6 + changes

text_processing/text_processing.tex

@@ -94,7 +94,7 @@
 %\input{ie_ner.tex} % end of 5th lecture + 6th lecture
 %\input{ie_relation_extraction.tex} % 7th lecture
 
-\input{tp_nn.tex}
+\input{textproc_nn.tex}
 
 \input{sa_extra_reading}
 

text_processing/text_processing_L6.tex

@@ -44,7 +44,7 @@
   Department of Computer Science\\
 }
 
-\date{November 26, 2019}
+\date{December 3, 2019}
 
 \begin{document}
 

text_processing/text_processing_L7.tex

+
+\documentclass[aspectratio=169,t,xcolor=table]{beamer}
+%\documentclass[t]{beamer}
+%\documentclass[handout,t]{beamer}
+
+\mode<presentation>
+{
+  %\usetheme{PaloAlto}
+%  \usetheme{Hannover}
+  \usetheme{informatics}
+  \useoutertheme{infolines}
+ % \setbeamercovered{transparent} % or whatever (possibly just delete it)
+}
+
+\setbeamertemplate{navigation symbols}{}
+
+\setlength{\extrarowheight}{3pt}
+
+\input{../LatexColors.incl.tex}
+\input ../mycolors.tex
+\input ../macros.tex
+\input ../macros_en.tex
+\input ../macros_beamer.tex
+
+\usepackage{setspace}
+
+%\setbeamercovered{transparent}
+
+\usepackage[absolute,showboxes,overlay]{textpos}
+%\TPshowboxestrue % commenter une fois fini
+\TPshowboxesfalse % décommenter pour faire disparaitre les boites
+\textblockorigin{10mm}{10mm} % origine des positions
+
+
+% This is only inserted into the PDF information catalog. Can be left out. 
+\subject{COM3110/Text Processing}
+\title[]{[Text Processing]\\ Deep Learning}
+
+\author[L. Barrault]{Loïc Barrault}
+
+\institute[University of Sheffield]
+{
+  l.barrault@sheffield.ac.uk \\
+  Department of Computer Science\\
+}
+
+\date{December 3, 2019}
+
+\begin{document}
+
+\begin{frame}
+  \titlepage
+\end{frame}
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%\input{textproc_introduction}
+\input{textproc_nn.tex}
+%\input{extra_reading}
+
+
+% and kill the abominable icon
+\setbeamertemplate{bibliography item}{}
+
+\begin{frame}[allowframebreaks]
+        \frametitle{References}
+%        \bibliographystyle{amsalpha}
+%        \bibliographystyle{apalike}
+        \bibliographystyle{IEEEtran}
+{\footnotesize        \bibliography{refs,refs_sa} }
+\end{frame}
+
+
+\end{document}
+
+

text_processing/tp_nn.tex → text_processing/textproc_nn.tex

@@ -5,7 +5,7 @@
 
 \vfill
 \centering
-\Huge{\edinred{Text processing\\Deep Learning}}
+\Huge{\edinred{[Text processing]\\Deep Learning}}
 
 \end{frame}
 
@@ -59,45 +59,51 @@ Hebb: \myemph{``Neurons that fire together, wire together''}
 \frametitle{The perceptron}
 
 \textbf{Perceptron}: computing unit loosely inspired by the biological neuron
-\vfill
 
-\centerline{
-\begin{tabular}[c]{cc}
-  \begin{tabular}[c]{c}
-  \includegraphics[width=0.4\textwidth]{figures/BpNeurone}
-  \end{tabular}
-  \hspace*{-1cm}
-  &
+\begin{columns}
+\begin{column}{.5\textwidth}
+\begin{center}
+	\includegraphics[width=0.6\textwidth]{figures/BpNeurone}
+\end{center}
+\end{column}
+\begin{column}{.5\textwidth}
+\begin{center}
   \begin{tabular}[c]{rl}
-    input: & $x_i$ \\
-    weights: & $w_i$ \\
+    \textbf{input}: & $\vx = \{x_i\}$ \\
+    \myemph{weights}: & $\vw = \{w_i\}$ \\
     threshold: & $s$ \\
     activity: & $\displaystyle a = \sum_i w_i x_i + s$ \\
-    output: & $y=f(a)$ \\
-  activation function: & $f=threshold(a)$
+    \myemph{activation function}: & $f=\text{threshold function}$ \\
+    \textbf{output}: & $\hat{y}=f(a)$ \\
   \end{tabular}
-\end{tabular}
-}
-\vfill
+\end{center}
+\end{column}
+\end{columns}
+\vspace{.5cm}
+Training method: change the weights $\vw$ if a training example $\vx$ is misclassified as follows:
+\begin{itemize}
+	\item[]   $\hat{\vw}^{new} = \hat{\vw}^{cur} + \hat{\vx} . y $ ~~~ with ~~~ $ y \in \{+1, -1\}$
+\end{itemize}
+	
 \end{frame}
 
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 \begin{frame}
 \frametitle{The Perceptron and the logical functions}
-
+\hspace{1cm}
 \begin{tabular}[t]{c}
-  y = a OR b \\[10pt]
+  y = a OR b \\[5pt]
   \includegraphics[height=0.7\textheight]{figures/or}
 \end{tabular}
-\hfill
+\hspace{1cm}%
 \begin{tabular}[t]{c}
-  y = a AND b \\[10pt]
-  \includegraphics[height=0.7\textheight]{figures/and}
+  y = a AND b \\[5pt]
+  \includegraphics[height=0.69\textheight]{figures/and}
 \end{tabular}
-\hfill
+\hspace{1cm}%
 \begin{tabular}[t]{c}
-  y = a XOR b \\[10pt]
-  \includegraphics[height=0.7\textheight]{figures/xor}
+  y = a XOR b \\[5pt]
+  \includegraphics[height=0.73\textheight]{figures/xor}
 \end{tabular}
 \end{frame}
 
@@ -125,13 +131,27 @@ y_i^{c} & = & f \left(\sum_j w^{c-1}_{ij} ~ y_j^{c-1}\right) \\
 \Ra\ \myemph{propagation} of the input $\vx$ towards the output $\vy$
 \end{frame}
 
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\begin{frame}
+\frametitle{How to train a multilayer perceptron?}
 
 
+\begin{block}{\center \textbf{Backpropagation}}
+\begin{center}
+\Large{Backward propagation of errors}
+\end{center}
+\end{block}
 
+\begin{itemize}
+\item What error? \Ra\ Error function depending on the task
+\item Estimating a real value: \ra\ 
+\end{itemize}
 
 
 
 
+\end{frame}
+
 
 
 
@@ -151,6 +171,15 @@ y_i^{c} & = & f \left(\sum_j w^{c-1}_{ij} ~ y_j^{c-1}\right) \\
 
 
 
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\begin{frame}
+\frametitle{TITLE}
+
+\end{frame}
+
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 \begin{frame}
 \frametitle{Deep Learning for Sentiment Analysis}