lukas/bachelor-thesis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Releases Activity

improve introduction and overall structure
All checks were successful
Build LaTeX Document / build (push) Successful in 10m41s
Details

Browse Source
This commit is contained in:
Lukas Heiligenbrunner
2024-10-02 17:42:29 +02:00
parent a5494ece51
commit f2ddf05333
2 changed files with 10 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
src/implementation.tex
View File

@@ -1,5 +1,10 @@
\section{Implementation}\label{sec:implementation}
\subsection{Experiment Setup}\label{subsec:experiment-setup}
% todo
todo setup of experiments, which classes used, nr of samples
kinds of experiments which lead to graphs
\subsection{Jupyter}\label{subsec:jupyter}
To get accurate performance measures the active-learning process was implemented in a Jupyter notebook first.

10
src/introduction.tex
View File

@@ -3,13 +3,13 @@
Anomaly detection has especially in the industrial and automotive field essential importance.
Lots of assembly lines need visual inspection to find errors often with the help of camera systems.
Machine learning helped the field to advance a lot in the past.
PatchCore and EfficientAD are algorithms trained only on good data and then detect anomalies.
The problem is they need a lot of training data and time to train.
PatchCore and EfficientAD are state of the art algorithms trained only on good data and then detect anomalies within unseen (but similar) data.
One of their problems is the need of lots of training data and time to train.
Few-Shot learning might be a suitable alternative with essentially lowered train time.
In this thesis the performance of 3 Few-Shot learning algorithms will be compared in the field of annomaly detection.
In this thesis the performance of 3 Few-Shot learning algorithms will be compared in the field of anomaly detection.
Moreover, few-shot learning might be able not only to detect anomalies but also to detect the anomaly class.
The goal of this practical work is to test active learning within a simple classification task and evaluate its performance.
\subsection{Research Questions}\label{subsec:research-questions}
\subsubsection{Is Few-Shot learning a suitable fit for anomaly detection?}
@@ -28,4 +28,4 @@ How does it compare to PatchCore and EfficientAD?
I've tried different distance measures $\rightarrow$ but results are pretty much the same.
\subsection{Outline}\label{subsec:outline}
todo