The first «quantum programmer» of Zurich University of Applied Sciences
Under the supervision of Prof. Dr. Kurt Stockinger and Prof. Dr. Rudolf Füchslin, Alexandros Soultanis is the first student at ZHAW who works on quantum computing.
Quantum computers obey the laws of quantum physics and are much better suited to solving certain problems than classical computers. Unlike classical computers that work with bits (representing either 0 or 1), quantum computers operate on quantum bits, so-called qubits. A qubit represents a (quantum mechanical) mixed state between 0 and 1. A register with two qubits then represents a mix of four bit sequences 00, 01, 10, 11, one with three qubits already eight bit sequences, etc. The power of quantum computers is based on the following two facts: First, a register of n qubits represents 2n ordinary bit strings. Second, a quantum mechanical operation is executed on all of these bit strings simultaneously. Thus, quantum computers are highly performant parallel computers. The major difficulty in using quantum computers, however, is that due to quantum effects any read-out of a qubit register irreversibly alters its content. This phenomenon is due to Heisenberg's uncertainty principle - named after the Nobel Prize winner Werner Heisenberg, a pioneer in quantum physics. One of the big challenges now is to design algorithms based on quantum mechanical processes such that the read-out problem described above nevertheless delivers (at least with great probability) correct results.
One of the first companies to build a quantum computer is the Canadian company DWave. In the early stages, their quantum computers were mainly used by NASA, Google and some US research laboratories. The programming of quantum computers was thus only possible for a small number of specialists who had access to the highly sought-after quantum computers. Recently, IBM, Google or Rigetti Computing have been working on their own quantum computers. These companies offer quantum simulators that can be programmed in Python. At IBM it is even possible to run the computer program on a real quantum computer with 14 qubits for free.
Programming quantum computers is very different from programming a classical computer. Without a basic understanding of quantum physics, it is not possible to write a correct program. Thus, the initial hurdle is still relatively high, and partly reminiscent of the beginnings of programming in assembler.
In his project work Alexandros Soultanis worked on a special algorithm, the Grover algorithm, which can be used for fast searching of databases. In the worst case, classical computers require n steps to find a certain value in an unsorted data set, where n is the number of entries in the database. The Grover algorithm solves the problem in "root n" steps and is therefore significantly faster. Alexandros implemented the algorithm, tested it on a quantum simulator, and ultimately executed it on a true quantum computer from IBM. Certain technical limitations currently only allow handling very simple example problems. However, the algorithm implemented by Alexandros is directly scalable and could one day use the full power of quantum computation on appropriate systems. This makes Alexandros the first ZHAW student to program a real quantum computer!
In the course of the bachelor thesis, the team of Alexander Soultanis, Rudolf Füchslin and Kurt Stockinger will work on improvements of the algorithm and investigate how simple machine learning algorithms can be programmed on a quantum computer.