Abstract:This work describes the selection approach and analysis of existing AutoML frameworks regarding their capability of a) incorporating Quantum Machine Learning (QML) algorithms into this automated solving approach of the AutoML framing and b) solving a set of industrial use-cases with different ML problem types by benchmarking their most important characteristics. For that, available open-source tools are condensed into a market overview and suitable frameworks are systematically selected on a multi-phase, multi-criteria approach. This is done by considering software selection approaches, as well as in terms of the technical perspective of AutoML. The requirements for the framework selection are divided into hard and soft criteria regarding their software and ML attributes. Additionally, a classification of AutoML frameworks is made into high- and low-level types, inspired by the findings of. Finally, we select Ray and AutoGluon as the suitable low- and high-level frameworks respectively, as they fulfil all requirements sufficiently and received the best evaluation feedback during the use-case study. Based on those findings, we build an extended Automated Quantum Machine Learning (AutoQML) framework with QC-specific pipeline steps and decision characteristics for hardware and software constraints.

What problem does this paper attempt to address?

The problems that this paper attempts to solve are: how to integrate quantum machine learning (QML) algorithms into the automated machine learning (AutoML) framework, and evaluate the scalability and capabilities of existing AutoML frameworks when dealing with different types of machine learning problems, especially for practical problems in industrial applications. Specifically, this paper focuses on the following aspects: 1. **Evaluation and Classification of AutoML Frameworks**: - Research existing AutoML frameworks and analyze whether they can support quantum machine learning (QML) algorithms. - Systematically select appropriate AutoML frameworks through a multi - stage, multi - criteria approach to achieve support for QML algorithms. 2. **Automation of Quantum Machine Learning (QML)**: - Build an extended automated quantum machine learning (AutoQML) framework, including quantum - computing - specific pipeline steps and decision - making features. - Solve optimization problems under the constraints of quantum - computing hardware and software. 3. **Practical Problems in Industrial Applications**: - Through multiple industrial use - case studies, verify the performance of the selected AutoML frameworks in solving practical problems. - Evaluate the performance of these frameworks on different types of machine learning problems, such as classification, regression, etc. 4. **Criteria for Framework Selection**: - Screen and evaluate AutoML frameworks according to hard and soft criteria. Hard criteria include open - source nature, Python interface support, active community, etc.; soft criteria involve the functionality of the framework, supported data types, optimizers, etc. - Eventually, Ray is selected as the low - level framework and AutoGluon as the high - level framework because they have good scalability and performance while meeting the above requirements. ### Main Contributions of the Paper - **Systematic Evaluation of Existing AutoML Frameworks**: Through detailed evaluation of multiple frameworks, determine which frameworks can support the integration of QML algorithms. - **Construction of AutoQML Framework**: Propose an extended AutoQML framework that can automatically select and configure QML algorithms in the AutoML process and optimize their hyper - parameters. - **Verification in Industrial Applications**: Through multiple industrial use - cases, verify the effectiveness and feasibility of the selected frameworks in practical applications. ### Summary This paper aims to enable AutoML frameworks to better support quantum machine learning algorithms through systematic evaluation and extension, thereby providing more efficient and more automated solutions for future quantum - computing applications.

Bringing Quantum Algorithms to Automated Machine Learning: A Systematic Review of AutoML Frameworks Regarding Extensibility for QML Algorithms

AQMLator -- An Auto Quantum Machine Learning E-Platform

A Brief Review of Quantum Machine Learning for Financial Services

Quantum Machine Learning: Bridging Quantum Computing and Artificial Intelligence

Towards an Automatic Framework for Solving Optimization Problems with Quantum Computers

Quantum Computing Enhanced Service Ecosystem for Simulation in Manufacturing

Quantum Machine Learning Architecture Search via Deep Reinforcement Learning

MAQA: A Quantum Framework for Supervised Learning

Applications of Quantum Machine Learning for Quantitative Finance

Quantum Computing for Automotive Applications: From Algorithms to Applications

Quantum-Assisted Simulation: A Framework for Developing Machine Learning Models in Quantum Computing

Application-Oriented Benchmarking of Quantum Generative Learning Using QUARK

Quantum machine learning on near-term quantum devices: Current state of supervised and unsupervised techniques for real-world applications

AutoQML: Automated Quantum Machine Learning for Wi-Fi Integrated Sensing and Communications

A Survey on Quantum Machine Learning: Current Trends, Challenges, Opportunities, and the Road Ahead

Quantum Machine Learning Predicting ADME-Tox Properties in Drug Discovery

A Roadmap for Automating the Selection of Quantum Computers for Quantum Algorithms

PO-QA: A Framework for Portfolio Optimization using Quantum Algorithms

Creating Automated Quantum-Assisted Solutions for Optimization Problems

Quantum Machine Learning for Software Supply Chain Attacks: How Far Can We Go?