Authors wishing to submit papers for inclusion in a Special Session must follow the IEEE QAI 2025 Submission Guidelines for preparing and submitting their papers. To make sure the papers will be included in the Special Session, please make sure you select the desired Special Session name in the Subject Areas. Special Session papers will be reviewed as other regular conference papers and those accepted will be included in the conference proceedings published by IEEE Xplore.

1. Advancements and challenges in integrating high-performance and quantum computing into future systems

Organizers: Marco Lapegna, Raffaele Montella

The integration of Quantum Computing (QC) and High-Performance Computing (HPC) represents a frontier in computational science, where the strengths of both paradigms can complement each other to address complex, computationally intensive problems. Quantum Computing promises unprecedented capabilities in solving certain classes of problems, such as optimization, simulation, and cryptography, leveraging quantum superposition and entanglement. However, it is still in its nascent stages, facing significant challenges related to qubit coherence, error rates, and scalability. On the other hand, High-Performance Computing has a long-standing history of solving large-scale computational problems using classical systems, with mature technologies such as parallel processing and distributed computing. HPC excels in handling vast datasets and high-throughput computations but may struggle with specific types of problems that could be better suited for quantum algorithms. This workshop aims to explore the potential relationship between QC and HPC by analyzing their respective strengths, limitations, and the potential benefits of their integration. By fostering collaboration between the two communities, we seek to outline a roadmap for a future where both technologies are harmonized to achieve breakthroughs in fields such as material science, cryptography, artificial intelligence, and complex system simulations. We invite researchers and practitioners to share their experiences, insights, and reflections on the following, but not exclusive, topics:

  •  Hybrid programming models and applications
  • Tools and environments for QC-HPC interfacing
  • Scheduling and run-time system support
  • Software architecture for QC-HPC hybrid systems
  • Software environment for simulation of QC-HPC hybrid systems
  • HPC for QC error mitigation and circuits design
  • QC as an acceleration paradigm for HPC
  • State of the art of QC technologies
  • Experience and case studies in QC applications

2. Practical Quantum Machine Learning in the NISQ Era

Organizers: Amir Pourabdollah, Manuel Cuéllar, José Manuel Soto-Hidalgo

Quantum Machine Learning (QML) continues to gain traction as a compelling area of research, but its real-world applicability remains challenged by the constraints of current quantum hardware. Although demonstrating a clear quantum speed or accuracy advantage over classical computing in real-world scenarios remains an open challenge in the Noisy Intermediate-Scale Quantum (NISQ) era, other forms of advantage — such as efficiency in resource allocation — can still be meaningfully explored. This special session focuses on practical QML approaches that operate within the capabilities of NISQ devices — particularly those methods that take innovative approaches to leverage a limited number of qubits and/or are robust to noise and decoherence — to implement effective and efficient QML. Therefore, the session aims to highlight resource-efficient and implementable QML techniques that show promise for deployment in realistic scenarios. We welcome contributions that include, but are not limited to:

  • QML algorithms optimised for low-qubit or near-term quantum hardware
  • Hybrid QML approaches that make an efficient and balanced use of quantum/classical in order to deliver a practical solution
  • Applications of QML to real-world domains such as energy, healthcare, or cybersecurity
  • Demonstrations of working advantageous QML models on quantum hardware or simulators
  • Benchmarks and performance comparisons with classical approaches

By bringing together researchers working on the frontlines of practical quantum learning, this session will promote discussion around how QML can offer tangible value in the NISQ era and help bridge the gap between theory and deployment.

3. Quantum Annealing and Its Related Hybrid Technologies

Organizers: Amir Pourabdollah, Amir Alizadeh

Quantum Annealing and variational quantum algorithms such as the Quantum Approximate Optimisation Algorithm (QAOA) have emerged as promising approaches to solving combinatorial and large-scale optimisation problems. This special session aims to bring together researchers and practitioners working at the intersection of quantum algorithms, classical optimisation, and emerging computing architectures to explore both theoretical and practical advances in quantum optimisation. While demonstrating advantage of such algorithms in terms of speed or accuracy remains a challenge for many real-world problems, other meaningful advantages such as energy efficiency, scalability, and resource-awareness can be investigated and demonstrated. These aspects are particularly relevant when designing algorithms and systems intended to operate effectively on current or near-term hardware. The session encourages submissions that explore:

  • New algorithmic developments in Quantum Annealing and QAOA
  • Resource-aware strategies tailored to NISQ devices
  • Hybrid quantum-classical optimisation workflows and hardware-aware implementations on platforms such as GPUs, FPGAs, or CPUs
  • Real-world case studies in areas such as energy, logistics, or scheduling
  • Comparative studies between classical optimisation approaches and quantum annealing-based approaches

We aim at sharing innovative methods, benchmarking results, and new directions in quantum optimisation with a special focus on quantum annealing methods and its related/hybrid technologies. This will encourage collaboration between algorithm developers, application domain experts, and hardware architects. We also aim at providing an opportunity to bridge theoretical foundations with practical implementation challenges, and showcasing quantum annealing technologies in impactful, measurable ways.

4. Pulsed Quantum Machine Learning

Organizers: Nicolino Lo Gullo, Francesco Cosco

Over the past few years, since quantum computers have become accessible to researchers, the development and benchmarking of quantum machine learning algorithms has emerged as a major area of focus. As with classical computing, two main paradigms are being explored: digital (gate-based) and analog. Among these, the digital approach has attracted significantly more attention, driven by algorithms like QAOA and VQE. However, current NISQ (Noisy Intermediate-Scale Quantum) devices still struggle delivering the sought quantum advantage through gate-based computing, despite the growing number of qubits. This presents a timely opportunity to explore alternative paradigms, ones that lean toward the analog model but retain key features of digital computing, such as gate operations and the encoding of information in the computational basis. Unlike traditional gate-based computing, where such pulses are predefined during calibration, in this approach they become an integral part of the algorithm itself. This not only expands the possibilities for encoding information, but also enables the emulation of novel classical computing paradigms, such as neuromorphic architectures, which promise to revolutionize the field of machine learning. This Special Session seeks to bring together experts in Quantum Machine Learning, Quantum Optimization, Quantum Computing Hardware, and Quantum Optimal Control to discuss emerging trends in utilizing quantum computing architectures at the pulse level. The focus is on designing and implementing quantum machine learning tasks by directly manipulating qubits in Quantum Processing Units (QPUs) using one- and two-qubit (and resonators where appropriate) control pulses. This shift is particularly timely given the upcoming deployment of various quantum computers across Europe as part of the EuroHPC initiative. These systems will provide access to pulse-level control, opening new possibilities for experimentation and algorithm design.

5. Quantum Reinforcement Learning – Theory and Applications

Organizers: Samuel Yen-Chi Chen, Joongheon Kim, Soohyun Park, Huan-Hsin Tseng, Hsin-Yi Lin, Kuan-Cheng Chen, Zhiding Liang 

QRL is an emerging interdisciplinary field at the intersection of quantum computing and reinforcement learning, offering the promise of accelerating decision-making processes and improving learning efficiency through quantum advantage. This Special Session aims to gather researchers and practitioners to explore recent theoretical developments, algorithmic frameworks, and practical applications in QRL. Key topics include the formulation of variational quantum policies, quantum-enhanced value function estimation, quantum exploration strategies, and distributed QRL architectures for quantum agents. The session will emphasize both foundational studies and application-driven research, covering use cases such as quantum control, financial modeling, communication networks, and optimization in quantum networks and energy grids. By providing a dedicated platform for this rapidly growing field, the session seeks to foster collaboration, identify critical challenges, and chart future research directions essential to advancing Quantum AI.

6. Quantum-inspired computational intelligence

Organizers: Ferdinando Di Martino, Barbara Cardone

In this special section we solicit innovative research papers that involve the application of quantum-inspired computational intelligence algorithms in various domains and address relevant topics such as:

  • Use of state superposition concept, which allows to explore multiple solutions simultaneously, favoring a greater diversity in the population in evolutionary algorithms.
  • Entanglement simulation, which introduces dependencies between variables in a nonlinear way, useful for capturing complex relationships in data.
  • Extension of classical evolutionary algorithms, such as GA and PSO, in which quantum concepts are used to increase the efficiency of convergence.
  • More effective optimization methods, which ensure faster convergence and escape local minima better than traditional ones.
  • Adaptability in complex contexts, which improves performance in nonlinear, noisy or dynamic scenarios, such as robotics, data mining and pattern recognition.

Topics for this call for papers include but are not restricted to:

  • Combinatorial optimization (e.g. NP-hard problems)
  • Classification and clustering (e.g. Quantum-Inspired SVM, Quantum Clustering)
  • Forecasting and time series
  • Computer visioned image processing
  • Pattern recognition in intelligent monitoring systems
  • Adaptive control and robotics