Modul

see partial achievement

Students have a systematic understanding of the issues surrounding the use of currently available quantum computers and of solution approaches from the field of hybrid quantum machine learning. They will be able to transfer these findings to other problems and, in particular, evaluate the efficiency and feasibility for different data sets in practice. In addition, they will be able to interpret and understand current research results in quantum machine learning with the knowledge you have acquired.

After successfully completing the course, students will be able to

• Explain the basic concepts, motivation and challenges of quantum computing and quantum machine learning as well as current hybrid approaches;
• Analyze problems, select suitable approaches, describe them mathematically and convert them into circuit notation and apply them;
• Design their own solutions based on the concepts learned in the lecture and evaluate their efficiency.

This module aims to familiarize students with the theoretical and practical aspects of the hybrid use of quantum circuits in classical machine learning algorithms. In the first part of the lecture, the necessary mathematical foundations of quantum systems and their representation by qubits and quantum circuits are summarized before the advantages and possibilities of quantum computing are demonstrated on the basis of known quantum algorithms. Finally, an overview of current hybrid approaches in the field of quantum machine learning (OML) and their possible applications and limitations is provided:

• Fundamentals and basic concepts
  - Theoretical and practical basics of quantum computing
  - Taxonomy of quantum machine learning
• Overview of QML algorithms, e.g.
  - Variational Quantum Eigensolver
  - Quantum Approximat Optimization Algorithm
  - Quantum Autoencoder
  - Quantum Convolutional Neural Networks
  - Quantum Generative Adversarial Neural Networks
  - Quantum Kernels
• Current challenges, e.g.
  - Noise
  - Barren Plateaus

The module focuses in particular on the applicability to today's quantum computers and the scalability of the approaches presented.

- Attendance of the lecture "Machine Learning 1 - Basic Methods" is recommended
- Attendance of the lecture "Introduction to Quantum Computing" is recommended
- Knowledge of linear algebra is recommended
- Programming skills in Python are helpful

- Lecture attendance: 23h (2 SWS x 15)
- Preparation and follow-up: 45h (2 x 2 SWS x 15)
- Exam preparation: 22h
- Total: 90h / 30 = 3 credits