Exploring How to Handle Classical Data in Quantum Models

Quantum computing has garnered increasing interest from researchers, businesses, and the public in recent years. The term “quantum” has become a buzzword that many use to capture attention. With the rise of this field, quantum machine learning (QML) has emerged as an intersection of quantum computing and machine learning. As someone interested in machine learning and passionate about math and quantum computing, I find the concept of quantum machine learning very appealing. However, as a researcher in the field, I am also somewhat skeptical about the near-term applications of QML.

Today, machine learning powers tools such as recommendation systems and medical diagnostics by finding patterns in data and making predictions. In contrast, quantum computing processes information differently by leveraging effects such as superposition and entanglement. The field of quantum machine learning explores this possibility and seeks to answer the question: can quantum computers help us learn from data more effectively? However, as with anything related to quantum computing, it’s important to set clear expectations. Quantum computers are currently flawed and incapable of running large-scale programs.

That said, they can provide a proof of concept for the utility of QML in various applications. Moreover, QML isn’t meant to replace classical machine learning. Instead, it looks for parts of the learning process where quantum systems might offer an advantage, such as data representation, exploring complex feature spaces, or optimization. With that in mind, how can a data scientist or machine learning engineer dip their toes into the realm of QML? Any machine learning algorithm (quantum or classical) requires data. The first step is always data preparation and cleaning.

This article focuses on QML workflows and data encoding techniques. Before we dive into the data, let’s briefly define what quantum machine learning is. At a high level, quantum machine learning refers to algorithms that use quantum systems to perform machine learning tasks, including classification, regression, clustering, and optimization. Most current approaches fall into what we call hybrid quantum-classical models, where classical computers handle data input and optimization while quantum circuits are part of the model.

A helpful way to think about this is: classical machine learning focuses on feature design, while quantum machine learning often emphasizes encoding features into quantum states. Since data can take many forms, QML workflows may look different depending on the type of input and algorithm. If we have classical data and a classical algorithm, that is our typical machine learning workflow. The other three options are where things get interesting.

The first option is to use quantum data with a quantum model. This is the most straightforward approach, where the input is already a quantum state. However, this workflow is still limited in practice due to challenges such as access to quantum data, state preparation and control, and measurement constraints. The second option is to use quantum data with classical algorithms. At first glance, this seems like a natural extension. If quantum systems can generate rich, high-dimensional data, why not use classical machine learning models to analyze it? In practice, this workflow is feasible but comes with an important limitation: a quantum system is described by a state that contains exponentially many amplitudes. However, classical algorithms cannot directly access this state. Instead, we must measure the system to extract classical information.