Visualization of hidden layers in convolutional neural networks – Bc. Jakub Hruška
Bc. Jakub Hruška
Master's thesis
Visualization of hidden layers in convolutional neural networks
Visualization of hidden layers in convolutional neural networks
Abstract:
Stejně jak v mnoha oborech vzrostla popularita strojového učení, vzrostla i složitost používaných modelů. Oblast vysvětlitelné umělé inteligence (XAI) se zabývá interpretací skrytých vnitřních procesů komplikovaných modelů. V této práci popisujeme jednu z větví XAI zvanou post-hoc metody. Pomocí nich prozkoumáváme vnitřní struktury konvolučních neuronových sítí, tzv. mapy rysů (feature maps). Dále …moreAbstract:
The popularity of machine learning raises in many fields, and so does the complexity of utilized models. Explainable Artificial Intelligence (XAI) aims to disclose the decision-making processes of complex models. This thesis provides an introduction to a branch of XAI called post-hoc methods. We explore internal representations of deep convolutional neural networks, so-called feature maps, using post …more
Language used: English
Date on which the thesis was submitted / produced: 14. 12. 2021
Identifier:
https://is.muni.cz/th/huq0h/
Thesis defence
- Date of defence: 1. 2. 2022
- Supervisor: doc. RNDr. Tomáš Brázdil, Ph.D.
- Reader: RNDr. Filip Lux
Citation record
Full text of thesis
Contents of on-line thesis archive
Published in Theses:- světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatikyMasaryk University
Faculty of InformaticsMaster programme / field:
Artificial intelligence and data processing / Machine learning and artificial intelligence
Theses on a related topic
-
Modelling small-RNA binding using Convolutional Neural Networks
Eva Klimentová -
Automatic Recognition of User Interface States Using Convolutional Neural Networks
Klára Petrovičová -
Segmentation of Dense Cell Populations using Convolutional Neural Networks
Filip Lux -
Interpretation techniques for deep neural networks in digital histopathology
Martin Krebs -
Tomographic back-projection of either sparse or low-quality projection views, based on convolutional neural networks (CNN)
Payal JAIN -
Image Analysis Using Machine Learning Models
Norbert Komiňák