wip

thesis/Main.bbl

@@ -1591,6 +1591,80 @@
       \verb http://dx.doi.org/10.1109/5.726791
       \endverb
     \endentry
+    \entry{ef_concept_source}{article}{}{}
+      \name{author}{8}{}{%
+        {{hash=cde98454252ce53a9838df6015a87166}{%
+           family={Ye},
+           familyi={Y\bibinitperiod},
+           given={Min},
+           giveni={M\bibinitperiod}}}%
+        {{hash=99c2139712828885b388f7f7e83f4134}{%
+           family={Nie},
+           familyi={N\bibinitperiod},
+           given={Jie},
+           giveni={J\bibinitperiod}}}%
+        {{hash=8b4ed7eebb280dc2d0df1bec91e27c18}{%
+           family={Liu},
+           familyi={L\bibinitperiod},
+           given={Anan},
+           giveni={A\bibinitperiod}}}%
+        {{hash=9f4446a6d583221d3d173f806ecf8627}{%
+           family={Wang},
+           familyi={W\bibinitperiod},
+           given={Zhigang},
+           giveni={Z\bibinitperiod}}}%
+        {{hash=ebcb470ee777cbc56984a83379d29819}{%
+           family={Huang},
+           familyi={H\bibinitperiod},
+           given={Lei},
+           giveni={L\bibinitperiod}}}%
+        {{hash=d761d577d421d6a0566473ae6b8b342f}{%
+           family={Tian},
+           familyi={T\bibinitperiod},
+           given={Hao},
+           giveni={H\bibinitperiod}}}%
+        {{hash=2e463ab20a44bc493252994ca77f0fca}{%
+           family={Song},
+           familyi={S\bibinitperiod},
+           given={Dehai},
+           giveni={D\bibinitperiod}}}%
+        {{hash=eeea6461d631d4e9cb07d2abc2de6885}{%
+           family={Wei},
+           familyi={W\bibinitperiod},
+           given={Zhiqiang},
+           giveni={Z\bibinitperiod}}}%
+      }
+      \list{publisher}{1}{%
+        {Frontiers Media SA}%
+      }
+      \strng{namehash}{0fca66725a9966a967fc7893b180ddef}
+      \strng{fullhash}{0e37676c60146890b0c3819a1c8e441b}
+      \strng{fullhashraw}{0e37676c60146890b0c3819a1c8e441b}
+      \strng{bibnamehash}{0e37676c60146890b0c3819a1c8e441b}
+      \strng{authorbibnamehash}{0e37676c60146890b0c3819a1c8e441b}
+      \strng{authornamehash}{0fca66725a9966a967fc7893b180ddef}
+      \strng{authorfullhash}{0e37676c60146890b0c3819a1c8e441b}
+      \strng{authorfullhashraw}{0e37676c60146890b0c3819a1c8e441b}
+      \field{sortinit}{7}
+      \field{sortinithash}{108d0be1b1bee9773a1173443802c0a3}
+      \field{labelnamesource}{author}
+      \field{labeltitlesource}{title}
+      \field{issn}{2296-7745}
+      \field{journaltitle}{Frontiers in Marine Science}
+      \field{month}{8}
+      \field{title}{Multi-Year ENSO Forecasts Using Parallel Convolutional Neural Networks With Heterogeneous Architecture}
+      \field{volume}{8}
+      \field{year}{2021}
+      \verb{doi}
+      \verb 10.3389/fmars.2021.717184
+      \endverb
+      \verb{urlraw}
+      \verb http://dx.doi.org/10.3389/fmars.2021.717184
+      \endverb
+      \verb{url}
+      \verb http://dx.doi.org/10.3389/fmars.2021.717184
+      \endverb
+    \endentry
   \enddatalist
 \endrefsection
 \endinput

thesis/Main.tex

@@ -65,6 +65,7 @@
 % \draftcopyName{ENTWURF}{160}
  
 \usepackage{xcolor}
+\usepackage{xfrac}
 \usepackage{booktabs}
 \usepackage{multirow}
 \usepackage[colorinlistoftodos]{todonotes}
@@ -1067,11 +1068,11 @@ The decoder network (see figure~\ref{fig:setup_arch_lenet_decoder}) mirrors the
 
 Even though the LeNet-inspired encoder proved capable of achieving our degradation quantification objective in initial experiments, we identified several shortcomings that motivated the design of a second, more efficient architecture. The most important issue concerns the shape of the CNN's receptive field (RF) which describes the region of the input that influences a single output activation. Its size and aspect ratio determine which structures the network can effectively capture: if the RF is too small, larger patterns cannot be detected, while an excessively large RF may hinder the network from learning to recognize fine details. For standard image data, the RF is often expressed as a symmetric $n \times n$ region, but in principle it can be computed independently per axis.
 
-\todo[inline]{RF concept figur}
+\fig{setup_ef_concept}{figures/setup_ef_concept}{Receptive fields in a CNN. Each output activation aggregates information from a region of the input; stacking layers expands this region, while kernel size, stride, and padding control how quickly it grows and what shape it takes. (A) illustrates slower, fine-grained growth; (B) shows faster expansion, producing a larger—potentially anisotropic—receptive field and highlighting the trade-off between detail and context. Reproduced from~\cite{ef_concept_source}}
 
-The RF shape's issue arises from the fact that spinning multi-beam LiDAR oftentimes produce point clouds posessing dense horizontal but limited vertical resolution. In our case this, this results in a pixel-per-degree resolution of approximately $0.99^{\circ}$/pixel vertically and $0.18^{\circ}$/pixel horizontally \todo[inline]{double-check with calculation graphic/table}. Consequently, the LeNet-inspired encoder’s calculated receptive field of $16 \times 16$ pixels translates to an angular size of $15.88^{\circ} \times 2.81^{\circ}$, which is highly rectangular in angular space. Such a mismatch risks limiting the network’s ability to capture degradation patterns that extend differently across the two axes.
+The RF shape's issue arises from the fact that spinning multi-beam LiDAR oftentimes produce point clouds posessing dense horizontal but limited vertical resolution. In our case this, this results in a pixel-per-degree resolution of approximately $5.69\,\sfrac{pixel}{deg}$ vertically and $1.01\,\sfrac{pixel}{deg}$ horizontally. Consequently, the LeNet-inspired encoder’s calculated receptive field of $16 \times 16$ pixels translates to an angular size of $15.88^{\circ} \times 2.81^{\circ}$, which is highly rectangular in angular space. Such a mismatch risks limiting the network’s ability to capture degradation patterns that extend differently across the two axes.
 
-\todo[inline]{add schematic showing rectangular angular RF overlaid on LiDAR projection}
+%\todo[inline]{add schematic showing rectangular angular RF overlaid on LiDAR projection}
 
 %\todo[inline]{start by explaining lenet architecture, encoder and decoder split, encoder network is the one being trained during the main training step, together as autoencoder during pretraining, decoder of lenet pretty much mirrored architecture of encoder, after preprocessing left with image data (2d projections, grayscale = 1 channel) so input is 2048x32x1. convolutional layers with pooling afterwards (2 convolution + pooling) convolutions to multiple channels (8, 4?) each channel capable of capturing a different pattern/structure of input. fully connected layer before latent space, latent space size not fixed since its also a hyperparameter and depended on how well the normal vs anomalous data can be captured and differentiated in the dimensionality of the latent space}
 %\todo[inline]{batch normalization, relu? something....}

thesis/bib/bibliography.bib

@@ -549,6 +549,20 @@
 	author = {Lecun, Y. and Bottou, L. and Bengio, Y. and Haffner, P.},
 	year = {1998},
 	pages = {2278–2324},
+},
+@article{ef_concept_source,
+	title = {Multi-Year ENSO Forecasts Using Parallel Convolutional Neural
+	         Networks With Heterogeneous Architecture},
+	volume = {8},
+	ISSN = {2296-7745},
+	url = {http://dx.doi.org/10.3389/fmars.2021.717184},
+	DOI = {10.3389/fmars.2021.717184},
+	journal = {Frontiers in Marine Science},
+	publisher = {Frontiers Media SA},
+	author = {Ye, Min and Nie, Jie and Liu, Anan and Wang, Zhigang and Huang, Lei
+	          and Tian, Hao and Song, Dehai and Wei, Zhiqiang},
+	year = {2021},
+	month = aug,
 }