TU Wien:Applied Deep Learning VU (Pacha)

Aus VoWi
Zur Navigation springen Zur Suche springen
Ähnlich benannte LVAs (Materialien):

Daten[Bearbeiten | Quelltext bearbeiten]

Vortragende Horst EidenbergerAlexander Pacha
ECTS 3
Letzte Abhaltung 2021W
Sprache English
Mattermost applied-deep-learningRegisterMattermost-Infos
Links tiss:194077, eLearning
Zuordnungen
Bachelorstudium Medizinische Informatik
Masterstudium Data Science
Masterstudium Data Science
Masterstudium Visual Computing
Katalog Freie Wahlfächer


Inhalt[Bearbeiten | Quelltext bearbeiten]

2024WS:

The course covers the following topics:

  • Introduction to Deep Learning
  • Neural Networks, Optimization, and Backpropagation
  • Convolutional Neural Networks and Visual Computing
  • Recurrent Neural Networks
  • Deep Reinforcement Learning
  • Autoencoders and Deep Generative Models
  • Transformers
  • Preprocessing, Data Augmentation, Regularization, and Visualization
  • Explainable AI
  • Graph Neural Networks
  • Large Language Models (LLMs)

Ablauf[Bearbeiten | Quelltext bearbeiten]

2024WS:

  • Lectures: Pre-recorded and uploaded to the ILEA platform.
  • Q&A Sessions: Weekly live Zoom meetings for questions and discussions.
  • Projects: Individual projects to be developed throughout the semester, with final presentations in the last two sessions.

Benötigte/Empfehlenswerte Vorkenntnisse[Bearbeiten | Quelltext bearbeiten]

  • No strict prerequisites since students choose their own projects based on their skill level. However, basic programming skills are highly recommended.

Vortrag[Bearbeiten | Quelltext bearbeiten]

2024WS:

  • Language: English
  • Format: Pre-recorded lectures that can be watched at any time.
  • One of the best aspects is that the lectures are recorded fresh every year, ensuring they are up-to-date with the latest deep learning advancements, rather than reusing outdated materials.
  • Professor Pacha does a good job delivering the content with clear explanations, diagrams, and practical examples.

Übungen[Bearbeiten | Quelltext bearbeiten]

2024WS:

  • Every student chooses their own project and works on it throughout the semester, creating a small demo at the end.
  • Three main submission deadlines:
    1. Project Proposal – Short description of the idea.
    2. Project Submission – Upload code to GitHub.
    3. Final Demo & Report – Record a short YouTube video of presenting your demo and submit a 5-page report.
  • The last two sessions are dedicated to watching and discussing the final presentations.

Prüfung, Benotung[Bearbeiten | Quelltext bearbeiten]

noch offen

Dauer der Zeugnisausstellung[Bearbeiten | Quelltext bearbeiten]

2024WS: Grades were posted on 20.02.2025

Zeitaufwand[Bearbeiten | Quelltext bearbeiten]

2024WS: I spent 61 hours in total. This includes working on the project and watching the weekly lectures. Grade obtained: 1

Unterlagen[Bearbeiten | Quelltext bearbeiten]

noch offen

Tipps[Bearbeiten | Quelltext bearbeiten]

Create a well-defined pipeline from the beginning to avoid last-minute struggles

Highlights / Lob[Bearbeiten | Quelltext bearbeiten]

👏 No group work

👏 Up-to-date lectures – Content is refreshed yearly.

👏 Custom projects – You can work on something that truly interests you.

Verbesserungsvorschläge / Kritik[Bearbeiten | Quelltext bearbeiten]

noch offen