Principles of Catastrophic Forgetting for Continual Semantic Segmentation in Automated Driving

Deep learning excels at extracting complex patterns but faces catastrophic forgetting when fine-tuned on new data. This book investigates how class- and domain-incremental learning affect neural networks for automated driving, identifying semantic shifts and feature changes as key factors. Tools for...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Kalb, Tobias Michael
Format: E-Book
Sprache:Englisch
Veröffentlicht: KIT Scientific Publishing 2024
Schriftenreihe:Karlsruher Schriften zur Anthropomatik
Schlagworte:
Automated Driving
Automatisiertes Fahren
Catastrophic Forgetting
Computer science
Computing and Information Technology
Continual Learning
Deep Learning
Katastrophales Vergessen
Kontinuierliches Lernen
Mathematical theory of computation
Maths for computer scientists
Semantic Segmentation
Semantische Segmentierung
ISBN:3731513730, 9783731513735
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Deep learning excels at extracting complex patterns but faces catastrophic forgetting when fine-tuned on new data. This book investigates how class- and domain-incremental learning affect neural networks for automated driving, identifying semantic shifts and feature changes as key factors. Tools for quantitatively measuring forgetting are selected and used to show how strategies like image augmentation, pretraining, and architectural adaptations mitigate catastrophic forgetting.
ISBN:3731513730
9783731513735
DOI:10.5445/KSP/1000171902