Quantum computing for data science: a practical test

A real hype has spread around the topic of quantum computing. Technology can open up new horizons in data science and machine learning. But how much do I work today? Based on testing in three use cases, the authors examine the specific benefits for data scientists.

Quantum computing basics

Unlike classical computers, quantum computer algorithms are not based on qubits but on qubits. A bit can have either state 0 or 1. If you measure a set bit multiple times, you will always get the same result. This varies with qubits. In principle, as strange as that sounds, it can take the values ​​0 and 1 at the same time. If you measure a qubit value multiple times, the values ​​0 and 1 occur with a certain probability. In the initial state, this is usually one hundred percent for the value 0. Other probability distributions for qubits can be generated by overlapping states. This is made possible by quantum mechanics, which follows laws we don’t know in this form of our everyday environment.

The crucial advantage of a quantum computer lies in its possibilities. Classic computers are powerful at problems where you need one bottom line. In contrast, quantum computers are very good at dealing with probability and can calculate using several values ​​at the same time. If you perform a one-time operation on a qubit in a superimposed state, it will be applied to both 0 and 1. A qubit represents both states at the same time. The more bits included in the computation, the greater the advantage over a conventional computer. For example, a computer with three qubits can cover up to eight states (which corresponds to 2³) at the same time: the binary numbers 000, 001, 010, 011, 100, 101, 110 and 111.

There is general agreement in the scientific literature that quantum computers will help solve previously unsolvable problems, including in the fields of data science and artificial intelligence. However, there are currently no perfect quantum computers available. The current generation is called Noisy Intermediate-Scale Quantum (NISQ). These computers are limited in their number of bits and are prone to interference, and they are prone to noise. In 2021, the first companies were able to build quantum computers with more than 100 qubits, including IBM and QuEra Computing. But what is the practical benefit of this generation? This will be demonstrated in the following test, in which the authors implemented three use cases with the Qiskit and PennyLane frameworks and verified their practical suitability. Compared to alternatives like Cirq (Google) and Q# (Microsoft), IBM’s Qiskit framework offers very good documentation and the advantage of being able to run circuits on a real quantum computer for free.