Machine learning \& artificial intelligence in the quantum domain (1709.02779v1)

Abstract: Quantum information technologies, and intelligent learning systems, are both emergent technologies that will likely have a transforming impact on our society. The respective underlying fields of research -- quantum information (QI) versus ML and AI -- have their own specific challenges, which have hitherto been investigated largely independently. However, in a growing body of recent work, researchers have been probing the question to what extent these fields can learn and benefit from each other. QML explores the interaction between quantum computing and ML, investigating how results and techniques from one field can be used to solve the problems of the other. Recently, we have witnessed breakthroughs in both directions of influence. For instance, quantum computing is finding a vital application in providing speed-ups in ML, critical in our "big data" world. Conversely, ML already permeates cutting-edge technologies, and may become instrumental in advanced quantum technologies. Aside from quantum speed-up in data analysis, or classical ML optimization used in quantum experiments, quantum enhancements have also been demonstrated for interactive learning, highlighting the potential of quantum-enhanced learning agents. Finally, works exploring the use of AI for the very design of quantum experiments, and for performing parts of genuine research autonomously, have reported their first successes. Beyond the topics of mutual enhancement, researchers have also broached the fundamental issue of quantum generalizations of ML/AI concepts. This deals with questions of the very meaning of learning and intelligence in a world that is described by quantum mechanics. In this review, we describe the main ideas, recent developments, and progress in a broad spectrum of research investigating machine learning and artificial intelligence in the quantum domain.

Overview of Machine Learning and Artificial Intelligence in the Quantum Domain

The paper Machine learning & artificial intelligence in the quantum domain by Vedran Dunjko and Hans J. Briegel explores the interaction between quantum mechanics and the fields of ML and AI. The paper investigates how quantum technologies can leverage learning theories and algorithms, and conversely, how AI and ML can benefit from quantum enhancement. These studies manifest in multiple directions, which are categorized throughout the paper.

Key Areas of Exploration

1. Application of Classical Machine Learning to Quantum Physics:
   • Classical machine learning has been applied effectively to quantum experiments, aiding tasks like quantum state estimation and Hamiltonian metrology.
   • More complex systems such as Hamiltonian estimation, quantum control, and many-body physics benefit significantly from machine learning methods, illustrating the potential for machine-assisted learning and discovery in quantum laboratories.
2. Quantum Generalizations of Machine Learning Concepts:
   • The potential for expanding classical ML tasks into the field of quantum data (e.g., quantum state discrimination and pattern recognition) poses intriguing challenges and opportunities.
   • This section also speculates on the formulation of quantum process learning and extensions of reinforcement learning through quantum frameworks.
3. Quantum Enhancements for Conventional Machine Learning:
   • Researchers have investigated how quantum computing architectures can improve classical learning problems both in terms of sample and computational complexity.
   • The paper explores quantum computational learning theory and systems like neural networks and their potential advantages when augmented with quantum techniques.
4. Quantum Learning Agents and Elements of Quantum AI:
   • The concept of quantum-enhanced reinforcement learning and agent-environment paradigms are considered as stepping stones towards quantum general artificial intelligence.

Notable Achievements and Results

• Quantum PAC Learning: Demonstrates instances where quantum classification is significantly faster than classical counterparts, especially when leveraging quantum distribution, although some stipulations remain distribution-dependent.
• Quantum Neural Networks: Discusses the inherent challenges and potential implementations for meaningful quantization of neuronal models, exploring how quantum mechanics might influence their processing abilities.
• Adiabatic Quantum Optimization for Machine Learning: Focuses on specific instances in boosting techniques where optimization plays a crucial role, showcasing how quantum annealing can address specific ML tasks faster than classical strategies.

Implications and Future Directions

The cross-pollination between quantum information processing and AI/ML is leading to novel theoretical frameworks and practical applications. As quantum computing hardware becomes more prevalent, these theoretical discussions have the potential to evolve into tangible advancements that reshape ML/AI capabilities.

• Practical Applications:

Quantum enhancements in learning capacity and computational complexity point towards practical implementations that leverage quantum data oracles, particularly in big data and optimization-related tasks.

• Theoretical Developments:

Quantum enhancements in classical disciplines, such as machine learning, offer a rich avenue of research noting improvements in learning efficiencies and capacities, either through sample or computational complexity.

The paper makes a compelling case for continued investment and interest in the intersection of quantum technologies and machine learning frameworks — an endeavor likely to further extend the boundaries of both fields. The emphasis is on creating robust, scalable frameworks that utilize quantum computational insights to address classical and inherently complex machine learning tasks. Future developments in quantum hardware and theory will only make these applications more accessible and relevant.