Abstract: There are two different ways to combine the respective strengths of quantum computing and machine learning. On the one hand, the unique capabilities of quantum computation can be used to enhance existing and develop novel machine learning techniques (QC4ML). On the other hand, one can utilize state-of-the-art machine learning approaches to support the development of quantum computing hardware and software (ML4QC). This tutorial focuses on the latter.
A quantum computer should not be seen as a black box, but rather as a realization of a special computer architecture including several layers. At the lowest level, n-qubit registers are developed using different integration technologies. Those are driven by means of analog control and read-out electronics. Based on these two hardware layers, several software layers follow. With strong reference to the quantum technology incorporated, quantum execution and error correction are carried out, followed by a generalization using quantum instruction set architecture. The compiler and runtime layer pursue multiple tasks such as decomposing, optimizing, mapping, and scheduling given circuits. Finally, there is the top layer of quantum algorithms, in which relevant applications for solving complex problems are addressed.
In this tutorial, we will give an overview of current challenges and goals of machine learning-based activities related to the different layers of a full-stack quantum computer. Examples of topics covered are: quantum circuit characterization and clustering; quantum circuit transformation and optimization; quantum error correction; quantum circuit mapping and scheduling; hybrid and distributed quantum computation.
The tutorial will have both a strong theoretical and practical part. These will be interwoven as follows: first, the corresponding investigated layer of the quantum computer stack will be explained, whereas challenges are interactively motivated by means of smaller examples (i.e., a kind of mini-lecture with exercises). Afterwards, current scientific work in this particular domain is presented and explained (i.e., in a review article manner). However, the aim here is not to elaborate on the smallest details, but rather to convey state-of-the-art open questions and solution approaches. The tutorial will end with an embedding of the individual insights in an overall context that helps to describe future work in the field of machine learning for quantum computation.
Contents Level: The tutorial aims to be as self-contained as possible. For
this reason, no particular background knowledge of quantum
computation or machine learning is required. Each of the
addressed topics will be motivated by small, interactive examples,
followed by an explanation of the necessary basic
concepts. Based on this knowledge, the respective current
scientific work is presented and explained to the participants.
However, a basic understanding of computer science is of
The content will be distributed roughly as follows: 30%
beginner, 50% intermediate, 20% advanced.
Target Audience: The self-contained tutorial is intended to serve a broad
audience. We basically address everyone who is interested in
improving the development of a full-stack quantum computer
and who would like to experience a well-prepared introduction
to this topic. This includes researchers and students from the
fields of electrical and computer engineering, computer science,
applied physics and other domains, with or without prior experience in quantum computation and machine learning.
We would also like to bring interested practitioners from the industry closer to this emerging field of research. Any kind
of improvement of quantum computing (system) software can help increasing the potential usability of a quantum computer
in the company concerned.