Streaming IoT Data and the Quantum Edge - A Classic/Quantum Machine Learning Use Case

Aug 2, 2026

Upload by Ezra

With the advent of the Post-Moore era, data-intensive machine learning for distributed analytics faces escalating demands for speed and efficiency. The study examines quantum machine learning as a potential approach, while addressing practical barriers: classical-to-quantum data encoding, hyperparameter tuning, and integrating quantum hardware into edge-to-HPC distributed systems. Edge computing is investigated as an enabler for hybrid quantum-classical analytics. Preliminary quantum machine learning analytics are demonstrated on an IoT scenario, focusing on encoding and tuning challenges.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

1 Introduction

Post-Moore computing challenges and quantum machine learning potential

2 Quantum machine learning on the Quantum Edge

Data encoding and hyperparameter importance

Quantum Edge for hybrid edge systems

IoT scenario and experimental setup

2 Background

2.1 Quantum Computing

arXiv :2402 . 15542v1 [ cs .ET] 23 Feb 2024 Streaming IoT Data and the Quantum Edge: A Classic/Quantum Machine Learning Use Case Sabrina Herbst, Vincenzo De Maio, Ivona Brandic Vienna University of Technology, Vienna, Austria, [sabrina.herbst@tuwien.ac.at](sabrina.herbst@tuwien.ac.at) ,{vincenzo,[ivona](ivona}@ec.tuwien.ac.at)[}](ivona}@ec.tuwien.ac.at)[@ec.tuwien.ac.at](ivona}@ec.tuwien.ac.at) [Abstract.](Abstract. With the advent of the Post-Moore era)[ With the advent of the Post-Moore era](Abstract. With the advent of the Post-Moore era), [the scientific com](the scientific com)munity is faced with the challenge of addressing the demands of current data-intensive machine learning applications, which are the cornerstone of urgent analytics in distributed computing. Quantum machine learning could be a solution for the increasing demand of urgent analytics, providing potential theoretical speedups and increased space efficiency. However, challenges such as (1) the encoding of data from the classical to the quantum domain,(2) hyperparameter tuning, and (3) the integration of quantum hardware into a distributed computing continuum limit the adoption of quantum machine learning for urgent analytics. In this work, we investigate the use of Edge computing for the integration of quantum machine learning into a distributed computing continuum, identifying the main challenges and possible solutions. Furthermore, exploring the data encoding and hyperparameter tuning challenges, we present preliminary results for quantum machine learning analytics on an IoT scenario. 1 Introduction IoT data and machine learning have recently become the keystone of urgent computing [1,2] . Data from IoT devices can be processed by machine learning models to improve simulations of different scientific phenomena [3] . However, machine learning applications require a huge amount of data for training. While data are streamed from IoT devices, they need to be transferred, stored and processed under strict response time constraints [4], which requires a huge amount of storage, computational and network resources. Therefore, training of machine learning is often performed inside HPC facilities. However, we recently entered the Post-Moore era, which faces the scientific community with the challenges of scaling computing facilities beyond current limits, which are codified by Moore’s law and Dennard scaling. As a consequence, current HPC facilities struggle to scale with the increasing amount of data available, pushing the scientific community towards research in Post-Moore Computing to address this issue. Among different possibilities, quantum computing clearly stands out due to theoretical speedups and increased space efficiency. This is particularly true for quantum machine learning, whose potential benefits are (1) increased speed, (2) increased predictive performance, and (3) reduced amount of data needed for training [5] . 2 Sabrina Herbst, Vincenzo De Maio, Ivona Brandic Quantum machine learning requires adapting data from the classical to the quantum domain before training and inference, following a process that is often referred to as data encoding. The choice of data encoding method can significantly affect the performance and accuracy of a quantum machine learning model [6] . Also, the choice of hyperparameters is of capital importance for the performance and accuracy of trained models. In this work, we expand our idea of the Quantum Edge [7] by investigating the possibilities of applying Edge computing methodologies to enable fast and efficient quantum machine learning on hybrid systems. After defining the problem, we present our idea of the Quantum Edge and describe how it can be applied to the target scenario. We identify challenges and possible solutions, and provide some preliminary results of our work towards the goal of Quantum Edge, tackling data encoding and hyperparameter selection. We perform training and inference o

FAQs

What problem does the document address for machine learning in the Post-Moore era?

It targets the difficulty of scaling computing facilities beyond limits while supporting data-intensive machine learning, especially when large streaming IoT datasets must meet strict response-time constraints.

Why is quantum machine learning considered in this work?

Quantum machine learning is explored for theoretical speedups, improved predictive performance, and the possibility of reducing the amount of training data required.

What are the main challenges preventing broader adoption, and how does the paper approach them?

The paper highlights data encoding from classical to quantum form, hyperparameter tuning, and integrating quantum hardware into distributed systems. It investigates using edge computing within a hybrid distributed continuum and reports preliminary IoT-focused results.

Aug 2, 2026

Upload by Ezra