Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Machine Learning for Quantum Computing Specialists

Daniel Goldsmith, M M Hassan Mahmud

TL;DR

This paper provides a concise primer linking classical machine learning concepts to quantum computing contexts for specialists. It covers core ML techniques—supervised and unsupervised learning, kernel methods, PCA, deep learning, and GANs—and discusses how quantum feature maps and quantum kernels could enable new capabilities on NISQ devices. It emphasizes the current hardware limitations, such as high error rates and short decoherence times, and outlines the long horizon for fault-tolerant quantum computing to unlock scalable QML. By tying ML concepts to quantum notions with illustrative examples, the paper offers a practical entry point and signposts for further study.

Abstract

Quantum machine learning (QML) is a promising early use case for quantum computing. There has been progress in the last five years from theoretical studies and numerical simulations to proof of concepts. Use cases demonstrated on contemporary quantum devices include classifying medical images and items from the Iris dataset, classifying and generating handwritten images, toxicity screening, and learning a probability distribution. Potential benefits of QML include faster training and identification of feature maps not found classically. Although, these examples lack the scale for commercial exploitation, and it may be several years before QML algorithms replace the classical solutions, QML is an exciting area. This article is written for those who already have a sound knowledge of quantum computing and now wish to gain a basic overview of the terminology and some applications of classical machine learning ready to study quantum machine learning. The reader will already understand the relevant relevant linear algebra, including Hilbert spaces, a vector space with an inner product.

Machine Learning for Quantum Computing Specialists

TL;DR

This paper provides a concise primer linking classical machine learning concepts to quantum computing contexts for specialists. It covers core ML techniques—supervised and unsupervised learning, kernel methods, PCA, deep learning, and GANs—and discusses how quantum feature maps and quantum kernels could enable new capabilities on NISQ devices. It emphasizes the current hardware limitations, such as high error rates and short decoherence times, and outlines the long horizon for fault-tolerant quantum computing to unlock scalable QML. By tying ML concepts to quantum notions with illustrative examples, the paper offers a practical entry point and signposts for further study.

Abstract

Quantum machine learning (QML) is a promising early use case for quantum computing. There has been progress in the last five years from theoretical studies and numerical simulations to proof of concepts. Use cases demonstrated on contemporary quantum devices include classifying medical images and items from the Iris dataset, classifying and generating handwritten images, toxicity screening, and learning a probability distribution. Potential benefits of QML include faster training and identification of feature maps not found classically. Although, these examples lack the scale for commercial exploitation, and it may be several years before QML algorithms replace the classical solutions, QML is an exciting area. This article is written for those who already have a sound knowledge of quantum computing and now wish to gain a basic overview of the terminology and some applications of classical machine learning ready to study quantum machine learning. The reader will already understand the relevant relevant linear algebra, including Hilbert spaces, a vector space with an inner product.
Paper Structure (28 sections, 20 equations, 21 figures)

This paper contains 28 sections, 20 equations, 21 figures.

Table of Contents

  1. Introduction
  2. Quantum Computing has given you a good start to machine learning
  3. What is Artificial Intelligence?
  4. My first experience of AI - the VAT "expert system'
  5. An introductory example - linear regression
  6. Aim of the linear regression example
  7. Loading the data using pandas
  8. Machine Learning terminology used to describe the data
  9. Training and test data
  10. The hypothesis space
  11. The inductive bias
  12. Choosing and optimising a loss function
  13. Useful linear regression documentation
  14. Overfitting
  15. Clustering
  16. ...and 13 more sections

Figures (21)

  • Figure 1: How Artifical Intelligence, Machine Learning and Deep Learning fit together
  • Figure 2: The COMPAQ microcomputer - Thanks to the Norsk Tekisk Museum CC BY-SA 3.0
  • Figure 3: Plot of employee data against company turnover: the raw data
  • Figure 4: The inductive bias identifies the hypothesis space to be used from all possible hypothesis spaces
  • Figure 5: The loss function shown diagramatically
  • ...and 16 more figures