FS2024: 31088/61088 Deep Learning

This course provides an introduction to deep learning methods. These are modern methods in artificial intelligence (AI), which are today incorporated in most of the top-performing algorithms to solve data-driven problems. Deep learning is based on the machine learning paradigm, where rules are learned from examples, rather than being hard-coded. Most examples will be in computer vision applications, that is, about problems of object recognition, detection, and segmentation in images and videos. On a successful completion of this course you can expect to be able to develop intelligent software to automate routine labor, understand images and videos (but you should be able to work with other data too), and support basic scientific research.

The course will cover most of the contents of the Deep Learning book by Goodfellow et al. and also recent developments in the field.
Topics:
Review of Machine Learning,
Deep feedforward networks (anatomy of a neural network, gradient-based learning, back-propagation, regularization, optimization, training),
Convolutional Neural Networks,
Natural Language Processing,
Recurrent Neural Networks/Transformers,
Autoencoders,
Representation learning,
Generative learning.

Prerequisites: applied math fundamentals like linear algebra, probability and numerical optimization

Books
Deep Learning by Ian Goodfellow, Yoshua Bengio, Aaron Courville (available online at http://www.deeplearningbook.org)
Dive into Deep Learning by Zhang et al (available online at https://d2l.ai/)


Learning Outcomes
On successful completion of this course students are expected to:
1) be able to formalize data-driven tasks via neural networks
2) design neural networks to solve data-rich tasks
3) build datasets, tune and train neural networks with advanced deep learning libraries
4) understand the inner mechanisms of neural networks during training
5) analyze the performance of neural networks on tasks of interest

Students interested in large scale data driven methods for high dimensional data
Students with already a grasp of machine learning principles and an interest in developing programming skills in deep learning

Englisch

This work has all rights reserved by the owner.

Unbegrenzt – wenn online geschaltet

Sie können diesem Kurs direkt beitreten.

Unbegrenzt

Anmeldename
Vorname
Nachname
E-Mail