3rd Workshop on Quantum Computing @ INFN

3 Feb 2026, 02:00 → 6 Feb 2026, 13:30 Europe/Rome

Auditorium U12 - Guido Martinotti

Andrea Giachero (INFN Milano-Bicocca), Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

Overview

The third Workshop on Quantum Computing  @ INFN will be held at the University of Milano-Bicocca on 3-6 February 2026.  

Quantum computing potentially offers a paradigm shift for issues of interest to INFN, in areas ranging from quantum machine learning to event reconstruction and simulation for experiments, theoretical physics, and many others.

The Quantum Computing @INFN workshop series represents an opportunity for the nuclear and subnuclear community to come together and discuss, with the objectives of presenting ongoing activities, fostering the exchange of knowledge and experiences, and attracting researchers and technologists who wish to acquire or enhance their skills.

Venue
Università di Milano-Bicocca, Auditorium U12 - Guido Martinotti

The workshop is organized in collaboration with the Italian Institute of Nuclear Physics (INFN), the University of Milano-Bicocca, and the Bicocca Quantum Technologies (BiQuTe) Centre

3qcinfn_flyer.pdf

informativa-eventi-2026-en-QC@INFN_2026.pdf

lunch options conference.pdf

Tuesday 3 February

10:00 → 11:00 Registration

11:00 → 11:20 Institutional Welcome

Conveners: Dr Andrea Giachero (INFN Milano-Bicocca), Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

11:00 Welcome from the Rector of the University of Milano-Bicocca

Speaker: Prof. Marco Orlandi (University of Milano-Bicocca)

11:05 Welcome from the Director of the INFN Unit of Milano-Bicocca

Speaker: Tommaso Tabarelli De Fatis (Istituto Nazionale di Fisica Nucleare)

11:10 Welcome from the Director of the Bicocca Quantum Technologies (BiQuTe) Centre

Speaker: Angelo Enrico Lodovico Nucciotti (Istituto Nazionale di Fisica Nucleare)

11:15 Welcome from the INFN Executive Committee

Speaker: Oscar Adriani (Istituto Nazionale di Fisica Nucleare)

11:20 → 13:00 Strategy and Insight from INFN Commissions

Convener: Dr Andrea Giachero (INFN Milano-Bicocca)

11:20 Quantum Strategy: Europe, Italy and the Italian Quantum Alliance

We will present the Italian National Strategy for Quantum and the European framework within which it is going to be deployed. We will also present the newly born Italian Quantum Alliance.

Speaker: Elisa Ercolessi (Istituto Nazionale di Fisica Nucleare)

QUANTUM@INFN_03feb2026_2.pdf

11:45 Quantum Technologies @CSN5

The talk outlines the foundational principles that guide Technological and Interdisciplinary Research Commission (CSN5) in selection and funding of research activities in the field of quantum technologies. It will present the scientific evaluation criteria, strategic impact considerations, and priority areas that shape the Commission’s investment decisions. The presentation will then review the history of quantum-technology initiatives promoted and funded by the Commission in recent years, analyzing the evolution of funding strategies, the outcomes achieved, and future perspectives. The contribution will provide a comprehensive overview of policies supporting quantum research and their impact on the development of the field.

Speaker: Alberto Quaranta (TIFPA - University of Trento)

Workshop QC_Quaranta.pdf

12:10 Quantum Activities in CSN4

I will briefly review the activities concerning quantum science that are carried out inside CSN4. Also the future of quantum activities inside CSN4 will be briefly discussed.

Speaker: Giuseppe Degrassi (Istituto Nazionale di Fisica Nucleare)

degrassi_2026-02-03.pdf

12:35 Quantum Machine Learning in High-Energy Physics: insights from INFN CSN1 Activities

Quantum Machine Learning is emerging as a promising paradigm for advancing data analysis, simulation, and optimization in high energy physics and accelerator-based experiment. This presentation provides an overview of ongoing QML efforts within INFN High Energy Particle Physics with Accelerators Commission (CSN1) and across Italian institutions. I will highlight existing projects developing quantum algorithms for classification, anomaly detection, event reconstruction and simulation, and quantum-enhanced optimization, with connections to LHC and beyond-LHC physics programs. The talk aims to give a unified picture of the current landscape in CSN1, mapping current strengths and collaborations, and outline opportunities for synergy within INFN and with international initiatives.

Speaker: Stefano Giagu (Sapienza Università di Roma and Istituto Nazionale di Fisica Nucleare)

CSN1_QCatINFN_MIB.pdf

13:00 → 14:30 Lunch

14:30 → 16:15 Foundational studies

Conceptual and theoretical foundations of quantum mechanics and quantum information, including quantum correlations, entanglement, nonlocality, and the interpretation of quantum theory.

Convener: Elisa Ercolessi (Istituto Nazionale di Fisica Nucleare)

14:30 QuEra: Progress towards fault-tolerant quantum computing

QuEra’s neutral-atom architecture combines high coherence, flexible connectivity, and scalability. I will present recent progress toward fault-tolerant quantum computation, highlighting our logical processor based on reconfigurable atom arrays and advances in error correction demonstrated with academic partners. The talk will also touch on the integration of QuEra’s digital platform Gemini within HPC environments, our collaboration with NVIDIA on GPU–quantum co-design and AI-accelerated error decoding, and emerging applications within the industry.

Speaker: Tommaso Macrì (QuEra)

14:50 Entanglement Generation in Uniformly Accelerating Quantum Systems

The Unruh effect predicts that a uniformly accelerated observer in the Minkowski vacuum perceives a thermal bath, so that an accelerating detector behaves as an open quantum system interacting with an effective thermal environment. We exploited this perspective to study entanglement generation between two identical two-level atoms located at the same spacetime point and weakly coupled to the same external quantum fields, without making assumptions on the spacetime dimension, or on the number and nature of the fields. We derived a Gorini–Kossakowski–Lindblad–Sudarshan (GKLS) master equation for the reduced dynamics of the two detectors and from the corresponding asymptotic state we computed the concurrence in order to measure the entanglement. In particular, we characterised how this depends on the coupling tensor between the detectors and the fields, identifying the regimes in which the Unruh-induced open dynamics can generate or suppress entanglement. This framework points toward potential quantum-simulation architectures, where controllable platforms could provide accessible proxies for Unruh-induced open-system effects.

Speaker: Riccardo Acquaviva (Istituto Nazionale di Fisica Nucleare)

Talk Acquaviva.pdf

15:10 Complexity transitions in chaotic quantum systems: Nonstabilizerness, entanglement, and fractal dimension in SYK and random matrix models

Complex quantum systems—composed of many, interacting particles—are intrinsically difficult to model. When a quantum many-body system is subject to disorder, it can undergo transitions to regimes with varying non-ergodic and localized behavior, which can significantly reduce the number of relevant basis states.
It remains an open question whether such transitions are also directly related to an abrupt change in the system's complexity.
In this talk, I will study the transition from chaotic to integrable phases in several paradigmatic models, the power-law random banded matrix model, the Rosenzweig—Porter model, and a hybrid SYK+Ising model, comparing three complementary complexity markers—fractal dimension, von Neumann entanglement entropy, and stabilizer Rényi entropy.
For all three markers, finite-size scaling reveals sharp transitions between high- and low-complexity regimes, which, however, can occur at different critical points. As a consequence, while in the ergodic and localized regimes the markers align, they diverge significantly in the presence of an intermediate fractal phase.
Additionally, I will show that the stabilizer Rényi entropy is more sensitive to underlying many-body symmetries, such as fermion parity and time reversal, than the other markers.
As our results show, different markers capture complementary facets of complexity, making it necessary to combine them to obtain a comprehensive diagnosis of phase transitions.
The divergence between different complexity markers also has significant consequences for the classical simulability of chaotic many-body systems.

Speaker: Andrea Legramandi (Istituto Nazionale di Fisica Nucleare)

Legramandi.pdf

15:30 Multipartite entanglement of random states

In this talk, we investigate multipartite entanglement through the statistical properties of pure quantum states of $n$-qubits. By analyzing the distribution of purity among balanced bipartitions, we will compare Haar-typical states with the so called Hadamard states, the latter being characterized by equal weights in the computational basis. We analyze different classes of Hadamard states distinguished by their phase distributions. We will show how Hadamard states exhibit, on average, a higher degree of entanglement than Haar-typical states. In addition, we will show that a particular class of Hadamard states, characterized by real coefficients with alternating signs, referred to as ($\pm$)-states, appears especially relevant in the search for maximally multipartite entangled states, both for their structural simplicity and the increased likelihood of sampling highly entangled states. These results identify Hadamard states as a tractable yet promising class for exploring multipartite entanglement structures and advancing the characterization of maximally entangled quantum states

Speaker: Giorgia Trotta

TrottaMilano26.pdf

15:50 Geometric Aspects of Quantum Entanglement

We derive the Fubini–Study metric, which endows the manifold of multi-qubit quantum states with a Riemannian structure.
We then explore the deep relationship between this Riemannian structure—defined on the projective Hilbert space of the system—and the entanglement properties of the states it contains.
Thus, we derived a measure of entanglement, the Entanglement Distance (ED), a quantity preliminarily introduced in Ref. [PhysRevA.101.042129].

Speaker: Roberto Franzosi (DSFTA - Università di Siena)

Franzosi_Sem_INFN-Milano.pdf

16:15 → 16:45 Coffee / tea break

16:45 → 18:30 Foundational studies

Conceptual and theoretical foundations of quantum mechanics and quantum information, including quantum correlations, entanglement, nonlocality, and the interpretation of quantum theory.

Convener: Chiara Macchiavello (University of Pavia)

16:45 Statistical mechanics of multipartite entanglement in ‘Hadamard’ quantum state sunmanifolds

We investigate multipartite entanglement in a specific class of pure n-qubit quantum states, the uniform real-phased states referred to as Hadamard states, through a statistical mechanics framework, where the average bipartite purity maps onto an effective Hamiltonian of $2^n$ binary classical spins. In this correspondence, each Hadamard state uniquely corresponds to a classical spin configuration, while temperature emerges as a control parameter that continuously interpolates between randomly sampled states at high temperature and maximally multipartite entangled states (MMES) in the zero-temperature limit. Remarkably, the zero-temperature entropy directly counts the MMES within the class of Hadamard states. For small system sizes (n < 6), we perform exact enumeration, fully characterizing the energy landscape and associated thermodynamic observables, and validating known MMES counts. For larger systems (n = 6 and 7), where exact methods become computationally infeasible, we employ simulated annealing and tempering to efficiently sample the high-dimensional state space. Our analysis yields quantitative predictions of MMES counts and reveals how entanglement is statistically distributed across the Hadamard state manifolds. The results establish this family of states as an ideal platform for exploring multipartite entanglement through thermodynamic methods, offering both computational advances and physical insights into the structure of quantum entanglement in constrained Hilbert spaces.

Speaker: Paolo Scarafile (Napoli)

presentation workshop milano_paolo_scarafile.pdf

17:10 Gate-based enhancement in dark matter searches with superconducting qubits

Superconducting qubits have emerged as promising platforms for quantum sensing, including the detection of dark matter candidates that couple to photons, such as axions and hidden photons. Conventional haloscope experiments in the few-gigahertz range can be enhanced using transmon qubits to suppress dark count rates, either exploiting qubit excitation schemes [1] or Quantum Non-Demolition (QND) [2].
In our work, we consider an alternative detection method that leverages a direct energy exchange between the transmon qubit and a photon-coupled dark matter field [3]. A qubit resonant with the dark field undergoes Rabi oscillations, although the weak coupling results in low excitation probabilities within typical transmon coherence times. To overcome this limitation, we introduce a qubit gate-based nonlinear amplification method that effectively enhances the excitation probability, reducing the impact of readout errors and decreasing the number of measurements required for setting competitive exclusion limits on dark matter–electromagnetic coupling strength.
Our approach, compared to the standard method of sampling the Rabi-driven qubit excitation probability, offers up to a ten-fold speedup in hidden photon search experiments, differing from other enhancement schemes [4] by being easier to implement and compatible with Noisy Intermediate-Scale Quantum (NISQ) devices.
Through numerical simulations, we validate the effectiveness of our scheme, providing an analysis of the detector’s physical implementation. The design leverages a coplanar architecture, and we employ finite element and analytical methods to extract key Hamiltonian parameters, including coupling strengths and transmission spectra. The simplicity of our approach, combined with its substantial performance improvement, represents a significant advance in Dark Matter search via direct excitation, and more generally contributes to the field of low-power microwave detection.

Speaker: Roberto Moretti (University of Milano-Bicocca and INFN Milano-Bicocca)

moretti_roberto_slides.pdf

17:35 Average-computation benchmarking for local expectation values in digital quantum devices

As quantum devices progress towards a quantum advantage regime, they become harder to benchmark. A particularly relevant challenge is to assess the quality of the whole computation, beyond testing the performance of each single operation. I will introduce a scheme for this task that combines the target computation with variants of it, which, when averaged, allow for classically solvable correlation functions. Importantly, the variants exactly preserve the circuit architecture and depth, without simplifying the gates into a classically-simulable set. The method is based on replacing each gate by an ensemble of similar gates, which when averaged together form space-time channels [P. Kos and G. Styliaris, Quantum 7, 1020 (2023)]. I will introduce explicit constructions for ensembles producing such channels, all applicable to arbitrary brickwork circuits, and provide a general recipe to find new ones through semidefinite programming. The resulting average computation retains important information about the original circuit and is able to detect noise beyond a Clifford benchmarking regime. Moreover, I will provide evidence that estimating average-computation expectation values requires running only a limited number of different circuit realizations.

Speaker: Flavio Baccari

Benchmark_Baccari.pdf

18:00 Floquet Theory applied to Lindbladian Open Quantum Systems

Floquet theory is a general method that allows one to treat linear systems of differential equations with periodic generators. It decomposes the propagator associated with the dynamical evolution into two parts: one associated with the micromotion within a period, and another corresponding to the effective stroboscopic propagation after each period. This framework is widely used in closed quantum systems, where the unitarity of time evolution ensures that the effective stroboscopic part is always generated by an effective Hamiltonian. This property has given rise to a new branch of quantum physics: Floquet engineering. However, when the theory is applied to Lindbladian open quantum systems, specific particularities arise. This talk will address these issues.

Speaker: Antonio Sojo López (Universidad de Sevilla)

wqc1080p(Use as Backup).pdf

wqc1080p.html

Wednesday 4 February

09:00 → 10:45 Quantum Machine Learning

Algorithms and hybrid ML/quantum computing approaches, with applications to high-energy physics and data analysis in general.

Convener: Stefano Giagu (Sapienza Università di Roma and Istituto Nazionale di Fisica Nucleare)

09:00 Quantum Generative Models for Fragmentation Functions

Quantum computing provides a natural framework for generative modeling through sampling tasks with established complexity-theoretic advantages, yet standard parametrized-circuit approaches face persistent challenges in trainability and scalability. This talk reports recent progress on a differentiable quantum generative model (DQGM) based on quantum Chebyshev transforms, which enables post-training resolution scaling and efficient sampling without additional optimization. As a key application, we study fragmentation functions (FFs) of charged pions and kaons from single-inclusive hadron production in electron-positron annihilation. We learn the joint distribution of momentum fraction z and energy scale Q, and infer their correlations from the entanglement structure.

Speaker: Dr Michele Grossi (CERN)

Grossi_CERN_QCINFN26.pdf

09:20 Quantum Neural Network-enhanced Models For Fast Calorimeter Simulation In ATLAS

The increasing demands on simulation statistics for HL-LHC analyses challenge the scalability of traditional calorimeter simulation within all LHC collaborations. Fast simulation techniques based on machine learning have proven effective, yet further improvements may arise from quantum-inspired models.
We investigate the feasibility of integrating Quantum Neural Network (QNN) components into the ATLAS fast calorimeter simulation framework, focusing on two complementary approaches. The first one employs a hybrid quantum–classical pipeline where a QNN is used to learn and generate the latent space of calorimeter shower representations, subsequently integrated into a classical Generative Adversarial Network (GAN) for sample generation. The second approach explores a quantum-inspired generative model based on Invertible Neural Networks (INN), providing a reversible mapping between input kinematic variables and calorimeter observables, thereby enabling explicit likelihood evaluation and enhancing interpretability. We report on implementation details, performance and ideas for future development.

Speaker: Federico Andrea Guillaume Corchia (University of Bologna and INFN Bologna)

Quantum_FastCaloGAN_Corchia_INFN_Quantum_Milano.pdf

09:40 Quantum machine learning applications for classification and simulation tasks at the LHCb experiment

In this talk, an overview of quantum machine learning (QML) activities developed within the LHCb experiment will be presented, with particular emphasis on their applications to classification and simulation tasks. Examples include the jet flavour identification as well as quantum generative models aimed at accelerating the detector simulations. It will be shown how quantum algorithms can potentially compete with state-of-the-art classical machine-learning methods in the LHCb analysis workflow. Finally, prospects and challenges for deploying QML tools on near-term quantum hardware and their integration into future data processing strategies at LHCb will be discussed.

Speaker: Lorenzo Sestini (Istituto Nazionale di Fisica Nucleare)

quantum_lhcb_milano.pdf

10:00 Finding the Higgs boson with quantum machine learning

This talk investigates the application of Quantum Machine Learning (QML) simulated algorithms to classification problems in High Energy Physics, where distinguishing signal events from complex background processes is essential. We focus on the Higgs boson decay channel 𝐻→𝑍𝑍*→4ℓ, a clean yet statistically limited signature. The study evaluates the performance of QML models and examines their potential benefits in comparison to state-of-the-art classical machine learning approaches.

Speaker: Eric Ballabene (Istituto Nazionale di Fisica Nucleare)

Presentation.pdf

10:20 Boosted Objects in Vector Boson Scattering: from classical to quantum ML prospects

The study of Vector Boson Scattering (VBS) at LHC provides a unique window into the electroweak symmetry breaking mechanism. The polarization of the vector bosons enables precision tests of the heart Standard Model at the TeV scale, additionally sensibile to new physics. The highest experimental sensitivity can be achieved in the boosted regime, where the bosons are produced with large transverse momentum, and decay hadronically. However, the resulting collimated decay products are reconstructed as single large-R jets, posing a major challenge for event reconstruction and signal discrimination. In this talk we investigate quantum machine learning strategies for polarization reconstruction, particularly based on Quantum Convolutional Neural Network and Quantum Variational Algorithms.

Speaker: Fabrizio Napolitano (Istituto Nazionale di Fisica Nucleare)

QuantumWorkshopMilano26.pdf

10:45 → 11:15 Coffee / tea break

11:15 → 13:00 Quantum Machine Learning

Algorithms and hybrid ML/quantum computing approaches, with applications to high-energy physics and data analysis in general.

Convener: Stefano Giagu (Sapienza Università di Roma and Istituto Nazionale di Fisica Nucleare)

11:15 QChaiTEA: the quantum-inspired machine learning application of the QTEA library

Tensor Networks (TN), originally developed in the context of quantum many-body physics, have recently emerged as powerful and interpretable Machine Learning (ML) architectures. The objective of this talk is twofold. Firstly, provide an intuitive and practical perspective on TNML methods, starting from the seminal work that initially established its connection to supervised learning. Secondly, present QChaiTEA, the ML application of the QTEA framework, describing its design and structure. This library provides comprehensive coverage of the entire training and analysis process for a quantum-inspired ML model. It encompasses the embedding map from classical data to tensor network states, the modelling through different TN ansätze, the optimisation step, and the extraction of meaningful quantities to describe and interpret the trained model. The library has been designed to be highly flexible. It provides a variety of embeddings, including the conventional feature map in separable states and MERA convolutions. Furthermore, it implements two optimisation procedures: one being backpropagation and the other one inspired to the DMRG algorithm. Finally, an application in High Energy Physics is showcased. In this context, we compare TNML against classical ML in the jet tagging task for fast inference at the trigger level. We show that unique characteristics of TNs, such as computational efficiency and interpretability, play a crucial role.

Speaker: Alberto Coppi (Istituto Nazionale di Fisica Nucleare)

Coppi_QChaiTEA.pdf

11:35 Sn-based quantum machine learning for multi-object tracking

Identity management (IM) for multi-object tracking is the problem of evolving a belief state over track–object associations, accounting for mixing events and measurement uncertainties.

In the most general setting the problem state is described by a probability distribution over the $n!$ permutations of $n$ objects, whose exact representation and update are inefficient in classical computation. This forces machine learning methods to rely on approximations like the evolution of low-order marginal probabilities.

We investigate an efficient, unconstrained quantum machine learning approach based on non-abelian Fourier analysis over the symmetric group $S_n$.

Exploiting the efficient scaling of the quantum Fourier transform for $S_n$, we propose an iterative two-step quantum pipeline. The algorithm models identity mixing events by a diffusion step that acts in the spectral domain, smoothing the probability distribution via its spectral decomposition. The resulting state is then conditioned on identity observations through a Bayes update in the anti-transformed space.

We present the group theoretical formalism underlying the algorithm, and provide a first blueprint for the two main sub-routines, including scalability studies. Finally, we discuss the potential of this framework as a novel quantum machine learning approach to scalable multi-object tracking and related data-association tasks.

Speaker: Matteo Argenton (Istituto Nazionale di Fisica Nucleare)

11:55 Characterization and upgrade of a quantum graph neural network architecture for particle tracking

The LHC detectors are going to be upgraded to profit from the substantial increase of the LHC instantaneous luminosity. In particular, in the forthcoming High Luminosity phase of ATLAS and CMS the number of proton–proton interactions per beam crossing is expected to rise by a factor three, averaging 140–200 in future Runs 4 and 5. Similar upgrades are also planned for the LHCb and ALICE detectors in Run5. This increase in luminosity leads to larger, denser events, and, consequently, greater complexity in reconstructing charged particle tracks, thus motivating frontier research in new technologies.

Quantum computing and machine learning methods have proved themselves to be two of the most promising emerging computing technologies of the last years. At the intersection of these two fields, we upgrade and characterize a quantum graph neural network (QGNN) architecture for charged particle track reconstruction, evaluated on a simulated high luminosity dataset.

The model operates on a set of event graphs, each built from the hits generated in tracking detector layers by particles produced in proton collisions, performing a classification of the possible hit connections between adjacent layers. In this approach, the QGNN is designed as a hybrid architecture, interleaving classical feedforward networks with parametrized quantum circuits. We characterize the interplay between the classical and quantum components. We report on the principal upgrades to the original design, and present new evidence of improved training behavior, specifically in terms of convergence toward the final trained configuration. Finally, we provide an outlook on noisy-intermediate-scale QML methods for high energy physics.

Speaker: Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

Q@INFN26.pdf

12:15 Ultra-Low-Latency Tree Tensor Network Inference on FPGAs

Tensor Networks (TNs) are a powerful computational framework originally developed for the efficient representation and simulation of quantum many-body systems. In recent years, they have gained increasing attention in machine learning (ML), demonstrating competitive performance in supervised learning tasks compared to conventional models.
In this work, we investigate the suitability of Tree Tensor Networks (TTNs) for high-frequency, real-time inference by exploiting the low-latency and high-throughput capabilities of Field-Programmable Gate Arrays (FPGAs). We present and evaluate multiple hardware implementations of TTN-based classifiers, targeting both standard ML benchmarks and complex datasets arising from physics applications.
During training, a systematic analysis is performed to determine optimal bond dimensions and weight quantization schemes. This analysis is informed by entanglement entropy and correlation function measurements, which provide insight into the representational capacity required by the model and guide the selection of the TTN architecture.
Following training, the TTN models are mapped onto a dedicated FPGA accelerator integrated within a server environment, with inference fully offloaded to hardware. This enables highly efficient, fully pipelined execution, achieving substantial reductions in inference latency. As a demonstrative application, we deploy a TTN-based classifier for a High Energy Physics (HEP) use case, achieving sub-microsecond inference latency while maintaining competitive classification performance.
These results demonstrate the feasibility of deploying quantum-inspired TN models within Level-1 trigger systems of HEP experiments, satisfying the stringent latency and throughput requirements while preserving robust classification performance. This work establishes TNs as a promising paradigm for real-time decision-making on specialized low-latency hardware platforms.

Speaker: Lorenzo Borella (Istituto Nazionale di Fisica Nucleare)

Borella_QCWorkshop.pdf

12:35 FPGA-Accelerated Quantum Circuit Emulation for Efficient QML models Training

Quantum computers promise to revolutionize computation. However, nowadays, their usefulness is still strongly limited by hardware noise, limited device availability, and the considerable cost of simulating quantum circuits on classical machines. These challenges become particularly evident when considering real-world applications, such as training Quantum Neural Network (QNNs), where the required number of circuit evaluations can quickly become prohibitively large as qubit count and circuit depth increase.

To overcome this bottleneck, we introduce a hardware–software methodology that accelerates QNN training by offloading quantum-circuit execution to an efficient FPGA-based engine. This architecture uses deep pipelining and compact data representations to emulate quantum operations at high speed while the classical optimization loop remains in software. This co-designed workflow achieves at least a 1.4× end-to-end speedup over CPU-only execution, while gains increase for larger and deeper models, demonstrating that real-time emulation during learning is already feasible with today’s digital hardware.

This contribution is part of a broader research effort focused on practical quantum technologies across multiple fronts: quantum optimization with digital Ising machines; quantum benchmarking and compilation methodologies; and automated workflows for developing quantum solutions to real-world tasks, thus improving the accessibility of quantum computing. Our method will reduce the time and computational resources invested in the exploration of quantum models, enabling a scalable and affordable path towards QML and enriching a growing set of tools and methods that bridge the gap between physics-based devices, algorithmic innovation, and advanced computing architectures.

Speaker: Dr Deborah Volpe (Istituto Nazionale di Geofisica e Vulcanologia)

Workshop_Milano_04022026.pdf

Workshop_Milano_04022026.pptx

13:00 → 14:30 Lunch

14:30 → 16:15 Quantum Simulation

Digital and analog simulation of physical systems, including simulation
methods, algorithms, and practical use-cases.

Conveners: Francesco Pederiva (Istituto Nazionale di Fisica Nucleare), Francesco Pederiva (TIFP)

14:30 Nuclear dynamics on digital quantum computers

An accurate description of many-body dynamics is a major challenge for classical simulation techniques. Hamiltonian simulation on digital quantum computers offer the possibility of reducing the computational cost when tackling these problems. In this talk, I will discuss recent advances in the simulation of both nuclear and neutrino systems using quantum computers and show some results obtained with current generation devices.

Speaker: Alessandro Roggero (Istituto Nazionale di Fisica Nucleare)

slides_roggero_milano.pdf

14:55 Exploring complex quantum many-body dynamics with QuantumTEA

QuantumTEA is an open-source tensor-network library for simulating and analysing quantum many-body systems in regimes directly relevant to current quantum-simulation and quantum-computation experiments. By providing efficient tools for modelling both unitary and dissipative dynamics and for extracting physical observables and entanglement measures, QuantumTEA enables detailed exploration, certification, and benchmarking of NISQ-era platforms. Its flexible design allows large-scale simulations of non-equilibrium quantum dynamics, offering a versatile computational framework for interpreting and guiding experiments with current quantum technologies.

Speaker: Darvin Wanisch (University of Padova)

talk_milan_dw.pdf

15:20 Quantum Encoding of Polymer Physics

The sampling of dense ensembles of self-avoiding polymers provides a paradigmatically hard problem in statistical mechanics and computational physics, even when resorting to minimalistic lattice models. In a series of recent studies, we addressed the question whether these computational limitations can be overcome by quantum annealers or by resorting quantum-inspired algorithms. We show that the reformulation of the sampling problems in terms of q-bits leads to a completely new field-theoretic approach with unexpected deep connections with lattice gauge theory. This greatly reduces the computational cost of sampling compact and/or topologically non-trivial ensembles even when implemented on a classical computer. We also compare the result of implementing our algorithms on classical and quantum hardware.

Speaker: Prof. Pietro Faccioli (University of Milano-Bicocca and INFN Milano-Bicocca)

Quantum encoding of polymer physics.pdf

15:45 Collective neutrino oscillations in three flavors on qubit and qutrit processors

Collective neutrino flavor oscillations are of primary importance in understanding the dynamic evolution of core-collapse supernovae and subsequent terrestrial detection, but also among the most challenging aspects of numerical simulations. This situation is complicated by the quantum many-body nature of the problem due to neutrino-neutrino interactions, which demands a quantum treatment. An additional complication is the presence of three flavors, which often is approximated by the electron flavor and a heavy lepton flavor. In this work, we provide both qubit and qutrit encodings for all three flavors, and develop optimized quantum circuits for the time evolution and analyze the Trotter error. We conclude our study with a hardware experiment of a system of two neutrinos with superconducting hardware: the IBM Torino device for qubits and Advanced Quantum Testbed device at the Lawrence Berkeley National Laboratory for qutrits. We find that error mitigation greatly helps in obtaining a signal consistent with simulations. While hardware results are comparable at this stage, we expect the qutrit setup to be more convenient for large-scale simulations since it does not suffer from probability leakage into nonphsycial qubit space, unlike the qubit setup.

Speaker: Luca Spagnoli (Istituto Nazionale di Fisica Nucleare)

Milano_Spagnoli.pdf

16:15 → 16:45 Coffee / tea break

16:45 → 18:30 Technological aspects

Dedicated to hardware platforms, device technologies, and implementation issues in quantum information science.

Conveners: Francesco Pederiva (Istituto Nazionale di Fisica Nucleare), Francesco Pederiva (TIFP)

16:45 Quantum computing and security. A reasonable worry?

One of the practical usages of quantum computing has often been attached to Shor's algorithm, a quantum factorization algorithm that it is said to outperform all traditional implementation and sound the doom bell of all traditional public/private key encryption algorithms. But is there any need to actually worry about that at this moment?

This presentation will show the current state of implementations and explain why the hype can be safely ignored for the moment.

Speaker: Vincenzo Ciaschini (Istituto Nazionale di Fisica Nucleare)

shor.pdf

shor.pptx

17:05 The Qibo Ecosystem: Integrating Advanced QML with Real-Time Hardware Control

We present the latest advancements to the Qibo ecosystem, a full-stack platform for quantum algorithm development, simulation, and hardware execution. Recent developments focused on deeply integrating Quantum Machine Learning (QML) capabilities and real-time hardware control to accelerate research, prototyping, and deployment of quantum applications.

The dedicated Qiboml module introduces a flexible framework that simplifies the construction of complex, composite quantum-classical models and ensures interoperability with established classical ML libraries. This design minimizes the complexity involved in rapid prototyping and experimentation with varied model components, including different derivative rules, noise mitigation techniques, and automated on-the-fly chip re-calibration.

The calibration process itself is managed through the Qibocal module. Its protocol library now extends calibration and stability management beyond single-qubit operations to include support for entangling-gate schemes. These include the operation of flux-tunable transmons, which require the precise handling of pulse distortions. These mechanisms are now formally exposed within Qibolab to ensure their reliable implementation across diverse experimental setups, aligning with the available literature.

Speaker: Andrea Papaluca

The Qibo Ecosystem: Integrating Advanced QML with Real-Time Hardware Control

17:25 Chalmers automatic calibration, a software framework to build collaborations

Chalmers developed a new bring-up, calibration and characterization software solution that is used to study the chips produced by the group but that can be easily adapted to any chip. The software consists of two layers; a library of experiments for one and two-qubit operations, which is based on the publicly available Tergite-autocalibration and Bragi, the orchestration layer that performs chip-wide operations. This software solution is compatible with the Tergite framework, which is built to manage quantum processors.

The presentation will demonstrate the capabilities of the software including the capability of bring-up of single-qubit gates to median value above 99.8%, studies of crosstalk between qubits and possible mitigation strategies. We also developed a procedure for calibrating two qubit gates in the whole chip that can be performed in less than a day, while the daily calibration requires about 2 hours. Possible topics for collaboration will be highlighted throughout the presentation.

Speaker: Michele Faucci Giannelli (Chalmers University of Technology)

Michele Faucci Giannelli, Chalmers calibration.pptx

17:45 Hybrid workflow compilation for Quantum Natural Gradient optimizer in PennyLane

Hybrid quantum-classical algorithms rely on an efficient back-and-forth between executing quantum circuit operations and processing information on classical computers (e.g., variational optimization or mid-circuit measurements). In this context, compilation is becoming increasingly important to enhance the performance and flexibility of these hybrid pipelines.
During my Xanadu Residency in 2025, I worked on the implementation of a (q)jit-compatible Quantum Natural Gradient optimizer in PennyLane, enabling hybrid programs to run within a fully differentiable and compiled workflow. A key part of this effort was to develop a JAX-native version of the Quantum Natural Gradient optimization and its momentum-based extension. This required rethinking how the metric tensor and parameter updates are represented so that they can be expressed as JAX data structures and integrate smoothly with Catalyst, Xanadu lower-level compiler for hybrid quantum-classical programs.
The new optimizers now run as a streamlined and accelerator-friendly execution pipeline, enabling faster and more scalable training of variational quantum algorithms. These improvements highlight how the hybrid tools developed within the PennyLane open-source software ecosystem support more efficient workflows for near-term algorithms.

Speaker: Simone Gasperini (University of Bologna & INFN)

Gasperini - QC@INFN26.pdf

Thursday 5 February

09:00 → 10:45 Quantum Machine Learning

Algorithms and hybrid ML/quantum computing approaches, with applications to high-energy physics and data analysis in general.

Convener: Dr Andrea Giachero (INFN Milano-Bicocca)

09:00 Optimizing Complex Quantum Systems with Few Measurements

Gradient-based optimization is a key ingredient of variational quantum algorithms, with applications ranging from quantum machine learning to quantum chemistry and simulation. The parameter-shift rule provides a hardware-friendly method for evaluating gradients of expectation values with respect to circuit parameters, but its applicability is limited to circuits whose gate generators have a particular spectral structure. In this work, we present a generalized framework that, with optimal minimum measurement overhead, extends parameter shift rules beyond this restrictive setting to encompass basically arbitrary gate generator, possibly made of complex multi-qubit interactions with unknown spectrum and, in some settings, even infinite dimensional systems such as those describing photonic devices or qubit-oscillator systems. Our generalization enables the use of more expressive quantum circuits in variational quantum optimization and enlarges its scope by harnessing all the available hardware degrees of freedom.

Speaker: Leonardo Banchi (Istituto Nazionale di Fisica Nucleare)

Quantum@Milan26.pdf

09:25 Quantum noise modeling through Reinforcement Learning

This work presents a novel machine learning approach to characterize the noise impacting a quantum chip and emulate it during simulations. By leveraging reinforcement learning, we train an agent to introduce noise channels that accurately mimic specific noise patterns. The proposed noise characterization method has been tested on simulations for small quantum circuits, where it con- sistently outperformed randomized benchmarking, a widely used noise characterization technique. Furthermore, we show a practical application of the algorithm using the well-known Grover’s circuit and QFT circuit.

Speaker: Simone Bordoni (Sapienza universita di Roma)

Published paper

RLnoise (PDF).pdf

RLNoise (Sap Font).ppt

09:50 Exploring fixed points and eigenstates of quantum systems with reinforcement learning

We introduce a reinforcement learning algorithm designed to identify the fixed points of a given quantum operation. The method iteratively constructs the unitary transformation that maps the computational basis onto the basis of fixed points through a reward-penalty scheme based on quantum measurements. In cases where the operation corresponds to a Hamiltonian evolution, this task reduces to determining the Hamiltonian eigenstates. The algorithm is first benchmarked on random Hamiltonians acting on two and three qubits and then applied to many-body systems of up to six qubits, including the transverse-field Ising model and the all-to-all pairing Hamiltonian. In both cases, the algorithm is demonstrated to perform successfully; in the pairing model, it can also reveal hidden symmetries, which can be exploited to restrict learning to specific symmetry sectors. Finally, we discuss the possibility of post-selecting high-fidelity states even when full convergence has not been reached.

Speaker: María Laura Olivera Atencio (Universidad de Sevilla)

MLOA.html

10:15 Quantum reinforcement learning in the presence of thermal dissipation

A study of the effect of thermal dissipation on quantum reinforcement learning is performed. For this purpose, a nondissipative quantum reinforcement learning protocol is adapted to the presence of thermal dissipation. Analytical calculations as well as numerical simulation are carried out, obtaining evidence that dissipation does not significantly degrade the performance of the quantum reinforcement learning protocol for sufficiently low temperatures, in some cases being even beneficial. Quantum reinforcement learning under realistic experimental conditions of thermal dissipation opens an avenue for the realization of quantum agent to be able to interact with a changing environment, as well as adapt to it, with many plausible applications inside quantum technologies and machine learning [1, 2]

Speaker: Prof. Jesús Casado Pascual (Universidad de Sevilla)

Talk_JCP.pdf

10:45 → 11:15 Coffee / tea break

11:15 → 13:00 Technological aspects

Dedicated to hardware platforms, device technologies, and implementation issues in quantum information science.

Convener: Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

11:15 The Italian landscape of Quantum Computing

The presentation provides an overview of the Italian quantum-computing ecosystem through the lens of the activities carried out within Spoke 10 of the National Center for HPC, Big Data and Quantum Computing. It outlines the strategic objectives of Spoke 10 and its role in advancing quantum technologies in Italy, with emphasis on research directions that bridge quantum hardware, algorithms, and software stacks, as well as on efforts aimed at accelerating the adoption of quantum computing for industrial use cases. Within this context, the talk also presents a brief overview of the contributions of Politecnico di Milano, focusing on research in Machine Learning for Quantum Computing, including learning-based approaches for quantum circuit optimization.

Speaker: Paolo Cremonesi (Politecnico di Milano)

The Italian landscape of Quantum Computing.pdf

11:35 The 25 → 64 Qubits Superconducting Quantum Computer of the HPC National Center @Unina: Physics, Implementation, Operation & Hardware Evolution

Superconducting circuits have been up to now the most successful platform to build a quantum computer, being developed and used by major international companies since early stages. Napoli has a long-standing experience on superconducting electronics and on its key device i.e. the Josephson junction, and has assembled the first quantum computer in Italy “Partenope” based on a 25-qubits processor produced by Quantware aiming at a QPU with 64-qubits in the close future. Partenope has been the platform where to create a solid expertise for the characterization, calibration, benchmarking and implementation of subregisters of QPUs and to focus on all hardware aspects including control and read out. A full control of Partenope also due to a comprehensive handling of the physics behind including its noise issues, such as decoherence, error in the gate implementation, readout error, has allowed the run of various algorithms, paving the way more and more towards an open-source quantum computing platform. A profound understating of all the physics of the hardware has promoted progress in developing independent pathways with innovative solutions for novel quantum components. These range from a new type of qubit based on ferromagnetic Josephson junctions and a novel tunable qubits coupler to qubit readout based on Josephson digital phase detectors and to novel schemes of microwave demultiplexer. The path from the physics of the hardware to operation of a quantum computer will be the main focus of the contribution.

Speaker: Prof. Francesco Tafuri (Università di Napoli Federico II)

11:55 Full-Stack Quantum Engineering at Politecnico di Torino: The Lagrange System

The Politecnico di Torino has recently acquired Lagrange, a 5-qubit superconducting quantum computer based on the IQM Spark: the first of its kind installed in Italy. The system integrates a full-stack infrastructure that includes a cryogenic QPU operating at 18 millikelvin, a low-noise microwave control chain with tunable couplers and Purcell-filtered readout, and a multilayer software environment (Cortex, EXA and Station Control) for pulse-level access and circuit execution. Its on-premises configuration provides direct physical access to the hardware, a key feature enabling both advanced research and experimental education.
From a research perspective, Lagrange allows the development of compact physical models derived from the extraction of experimental observables, such as qubits crosstalk, gates characterization etc, supporting studies on qubit coherence and device-level parameter studies. The platform can handle relevant algorithms (e.g. QML, optimization) through the circuit knitting technique, where circuits are decomposed into sub-circuits suitable for 5-qubit execution. The system is being leveraged to design and test new FPGA-based qubit-control architectures, designed to react on timescales compatible with qubit coherence. Such low-latency, high-throughput control enables tighter feedback, more adaptive pulse correction, and hardware-level error-mitigation strategies.
In the educational field, Lagrange provides a unique opportunity for students of the new Master Degree in Quantum Engineering to engage directly with a real quantum processor. It supports the study of the full computational framework, from job scheduling to pulse calibration and circuit transpilation, and the testing and validation of quantum algorithms under realistic conditions. The ability to extract and analyze physical parameters also enables the creation of compact models, linking theoretical coursework with experimental practice and preparing the next generation of quantum engineers.

Speaker: Giovanna Turvani (Politecnico di Torino)

12:15 Quantum Computing at CINECA: Integrating a Quantum Accelerator into an HPC Environment

The rapid advancements in quantum computing technology have ushered in an era of unprecedented computational potential. CINECA, in collaboration with five other leading European countries, has been designated as a Hosting Entity for an EuroHPC Joint Undertaking quantum computer. This talk elucidates the role played by CINECA in the integration of quantum accelerators within High Performance Computing (HPC) environments, contributing to mark a crucial milestone towards achieving computing power beyond the exascale realm.
The presentation will delve into the intricacies of amalgamating quantum computational paradigms with traditional HPC architectures, addressing the critical challenges and opportunities that arise in this process. Key considerations encompassing hardware compatibility, resource allocation, and workflow optimization will be discussed, shedding light on the methodologies employed to harness the synergistic potential of quantum-accelerated computing within a well-established HPC ecosystem.
Furthermore, the talk will provide insights into the collaborative efforts and interdisciplinary expertise harnessed by CINECA's scientific community. The integration of quantum accelerators into the HPC infrastructure necessitates a convergence of quantum algorithm development, quantum hardware engineering, and classical computing expertise, thereby fostering a holistic approach towards achieving the “quantum advantage”

Speaker: Sara Marzella (Cineca)

INFN_CINECA_Marzella.pdf

12:35 Activities on Quantum Computing in the Salerno Area

Thanks to national (NQSTI) and local (Regione Campania) funding the University of Salerno is strongly investing in quantum computing and the related technologies. A large quantum computer, with more than 150 qubits, will be installed in the Salerno campus in the next months and will act as a propulsive force for the development of advanced quantum algorithms and applications tailored for working on reals quantum machines. At the same time a large investment has been done to realize a state-of-the-art foundry for realization of advanced quantum circuits. This is complemented by advanced design and modeling expertise and by mK cryostats for testing the developed devices. The effect of all these activities on the local and national panorama of quantum initiatives will be discussed in details.

Speaker: Prof. Sergio Pagano (Dipartimento di Fisica Università di Salerno and INFN GC Salerno)

INFN-MIB_Pagano_v5.pdf

13:00 → 14:30 Lunch

14:30 → 16:15 Technological aspects

Dedicated to hardware platforms, device technologies, and implementation issues in quantum information science.

Convener: Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

14:30 Italian Technological Platforms for Quantum Computing

The National Quantum Science and Technology Institute (NQSTI) has launched the QuantumFab, I will describe this network of centres of competence distributed across Italy where we have equipment and expertise to create systems, devices, and components for industry and for research throughout the country.

Speaker: Francesco Cataliotti (Università di Firenze - CNR)

Cataliotti_WsBicocca.pdf

14:55 Quantum Computers to accelerate the path to FTQC

IQM company presentation, Ecosystem, QPUs superconducting, topologies HPC-QC integration, Solutions for Research Centers and Universities

Speaker: Cosma Belli

FTQC - INFN Italy - Cosma.pdf

15:20 IBM - towards Quantum Advantage and Error Correction

IBM announced fundamental progress on its path to delivering both quantum advantage by the end of 2026 and fault-tolerant quantum computing by 2029.

IBM is unveiling IBM Quantum Nighthawk, its most advanced quantum processor yet and designed with an architecture for enhanced connectivity, to complement high-performing quantum software to deliver quantum advantage next year.

In a parallel path, IBM is rapidly delivering milestones towards building the world’s first large-scale, fault-tolerant quantum computer by 2029.
IBM has already demonstrated the breakthrough features that will be incorporated into Loon, including the introduction of multiple high-quality, low-loss routing layers to provide pathways for longer, on-chip connections (or “c-couplers”) that go beyond nearest-neighbor couplers and physically link distant qubits together on the same chip, as well as technologies to reset qubits between computations.

About the software point of view, Qiskit SDK v2.2 concerns Qiskit’s C API, which takes an important step toward building out our support for HPC environments with the introduction of a standalone transpiler function that is directly callable from C. With this capability in place, it is now possible to construct end-to-end quantum workflows that you can execute natively in C and other compiled languages integrated via the C API.

Speakers: Davide Moretti, Luca Cavallini, Vito Sammarco

IBM Quantum for INFN.pdf

15:45 VIO, an architecture for scaling superconducting quantum processors

Scaling quantum processors to the million-qubit regime is fundamentally constrained by the I/O density and fan-out limitations of planar architectures. We present VIO, a 3D architecture designed to overcome these scaling barriers. VIO employs a novel 90-degree packaging technique that connects the horizontal qubit plane to a vertical chipstack. This stack, composed of micromachined silicon substrates and superconducting circuits, provides dense, low-loss interconnects compatible with high-density flexible cabling.
We present key features of the VIO architecture, paving the way to 10k- and later million qubit systems: the co-integration of passive and active components (e.g., filters, parametric amplifiers) directly into the vertical stack, which minimizes signal loss, reduces routing complexity, and meets stringent cross-talk requirements; modular scaling by tiling quantum modules via low-loss chiplet-to-chiplet interconnects; and compatibility with chiplets of all superconducting qubit modalities.

Speaker: Alessandro Bruno (QuantWare)

16:15 → 16:45 Coffee / tea break

16:45 → 18:30 Quantum Simulation

Digital and analog simulation of physical systems, including simulation
methods, algorithms, and practical use-cases.

Convener: Dr Marco Cè (Università di Milano-Bicocca)

16:45 Lattice Gauge Theory simulations with Quantinuum Systems

Trapped-ion quantum computers based on the QCCD architecture are a scalable and high-fidelity quantum platform for the explorations of lattice gauge theory phenomena.
Quantinuum Systems, based on such technology, are among the most utilized quantum computers to study simplified models of QCD in 1 and 2 spatial dimensions.
They are a remarkable tool in improving the development of quantum algorithms for quantum simulations of out-of-equilibrium physics in lattice gauge theories, which are notoriously hard to simulate classically due to their increased computational complexity.
My talk will introduce the technology and the state-of-the-art simulation tools and results in the field of quant-hep, the combination of quantum tech and high energy physics.

Speaker: Enrico Rinaldi (Quantinuum)

17:05 Local fermion-to-qudit mappings

We present a family of local fermion-to-qudit mappings that exploit four-level systems to encode fermionic degrees of freedom with local qudit operators. By tailoring the mappings to both spinless and spinful fermions, we achieve lower qudit weight and shallower circuit constructions compared with standard fermion encodings such as the Jordan–Wigner transformation. The resulting representations preserve locality while maintaining compatibility with currently accessible multi-level hardware, particularly superconducting qudits. We benchmark the approach through Trotterized simulations of archetypal models—including the spinless t−V model and the two-dimensional Fermi–Hubbard model, demonstrating reductions in two-qudit gate counts without added computational overhead. These results indicate that qudit-based architectures provide a scalable and resource-efficient pathway for simulating interacting fermionic systems.

Speaker: Rodolfo Carobene (University of Milano-Bicocca and INFN Milano-Bicocca)

infn_workshop_rodolfo.pdf

17:25 Complex vs. Imaginary Time: Benchmarking Quantum-Inspired Annealing for the SK Spin Glass

Finding the ground state of many-body systems is a central challenge in statistical physics and combinatorial optimization. Hard optimization problems can be mapped onto spin-glass–like Hamiltonians whose ground-state configurations encode valid solutions. In this work, we introduce a quantum-inspired tensor network method to tackle this class of problems. Our approach extends standard quantum annealing by allowing for complex time evolution and projects the resulting dynamics onto a tensor network manifold using the time-dependent variational principle. We apply this method to the ground-state search of the paradigmatic Sherrington–Kirkpatrick (SK) spin-glass model. By comparing pure imaginary-time quantum annealing with complex-time annealing, we observe that the former exhibits superior performance in both accuracy and computational efficiency. We also analyze the scaling of computational effort and find that both the time to reach an accurate ground state and the required bond dimension grow only slowly with system size, indicating favorable scalability. Overall, our results indicate that tensor-network–based quantum-inspired optimization methods constitute a viable alternative within combinatorial optimization. Finally, we explore how convergence behavior may relate to structural indicators of instance hardness in typical combinatorial problems.

Speaker: Shima Amirentezari (University of Padova)

Shima_presentation_milan.pdf

17:45 Quantum state preparation with polynomial resources: Branched-Subspaces Adiabatic Preparation (B-SAP)

Quantum state preparation is a central challenge in quantum computation and quantum simulation, enabling the exploration of complex many-body phenomena in Quantum Mechanics, Quantum Field Theory, and Quantum Chemistry. Existing paradigms such as Variational Quantum Algorithms (VQAs) and Adiabatic Preparation (AP) offer viable pathways, but each suffers from intrinsic limitations—VQAs from barren plateaus and optimization overheads, and AP from stringent adiabaticity requirements and spectral bottlenecks.

In this work, we introduce Branched-Subspaces Adiabatic Preparation (B-SAP), a hybrid algorithm that merges the conceptual advantages of VQAs and AP by leveraging concepts from group theory and classical post-processing to enable efficient approximation of ground and excited states of many-body Hamiltonians. B-SAP generates a sequence of branched, symmetry-adapted subspaces that steer the adiabatic evolution and confine the search space, significantly reducing quantum resource requirements while relaxing the strict conditions typically required for the adiabatic theorem.

We validate the method on the one-dimensional XYZ Heisenberg model with periodic boundary conditions, benchmarking performance across a wide range of anisotropies and system sizes. Our results demonstrate accurate preparation of the low-energy eigenstates with circuit depths that scale polynomially with system size, highlighting B-SAP as a resource-efficient and scalable approach for quantum state preparation.

Speaker: Davide Cugini (Università di Pavia)

Davide_Cugini_INFN_2026_v3.pdf

18:05 Implementation in a quantum computer and quantum phase classification of an anharmonic Lipkin model.

The Lipkin-Meshkov-Glick model (Lipkin model for short) describes a two-level system of N particles interacting with a long-range interaction, which is further generalized by adding an anharmonicity term with an extra control parameter. It is of interest in nuclear physics as a simple non-trivial model with two symmetry configurations that give rise to two distinct quantum phases. In this work, the time-evolution of the anharmonic Lipkin model is implemented and executed in a quantum chip, measuring a dynamical correlation function. This correlation function is then employed to classify the corresponding quantum phase via a supervised machine learning algorithm trained on classical simulations of the simpler Lipkin model.

Speaker: ALVARO SAIZ CASTILLO (UNIVERSIDAD DE SEVILLA)

ALVARO SAIZ - Implementation in a quantum computer and quantum phase classification of an anharmonic Lipkin model..pdf

Friday 6 February

09:00 → 10:00 Quantum Machine Learning

Algorithms and hybrid ML/quantum computing approaches, with applications to high-energy physics and data analysis in general.

Convener: Dr Andrea Giachero (INFN Milano-Bicocca)

09:00 Quantum-Enhanced Fraud Detection: A Comparative Study on Real-World Financial Data

Financial fraud detection is a highly imbalanced and dynamic learning problem that challenges traditional machine learning models. Quantum machine learning (QML) offers new representational paradigms capable of exploring high-dimensional feature spaces in ways substantially different from classical systems, leveraging quantum superposition and entanglement. In this study, we present a comprehensive evaluation of quantum and hybrid QML architectures, specifically Quantum Autoencoders, Hybrid Diffusion Time Estimation, and Hybrid Variational Autoencoders, applied to real-world banking data encompassing millions of transactions. By combining deterministic preprocessing, feature selection, and balanced validation protocols, we assess the comparative advantages of quantum-enhanced inference under realistic fraud detection conditions. We implement quantum models via classical simulations. Results indicate that even if current NISQ hardware constraints limit scalability, quantum–classical hybrids can yield measurable improvements in anomaly sensitivity and calibration over standard baselines for specific data regimes. Classical models generally remain more consistent and robust overall. This study provides one of the first end-to-end empirical analyses of QML in operational financial contexts, illustrating both its feasibility and open research challenges.

Speaker: Andrea Cacioppo (Istituto Nazionale di Fisica Nucleare)

QFDbis.pdf

09:20 A new approach to rating scale definition with quantum optimization

In the field of finance, lenders must estimate the creditworthiness of loan applicants, which leads them to devise strategies for classifying borrowers. Among these strategies, banking groups identify the definition of a rating scale as a classification method regulated by national and international constraints. The definition of a rating scale can be described as a combinatorial optimization problem.

The mathematical formulation known as Quadratic Unconstrained Binary Optimization (QUBO) can address a wide variety of significant combinatorial optimization problems, including rating scale definition. It has been proven that this kind of formulation can benefit from quantum computing due to its equivalence to a quantum system described by an Ising Hamiltonian.

In the context of the ICSC – Centro Nazionale di Ricerca in HPC, Big Data and Quantum Computing, a collaboration project between INFN and Intesa Sanpaolo Bank aims to build a QUBO formulation to define a rating scale for classifying borrowers. We present the final results of this project, starting from the definition of the financial constraints up to its implementation and testing using different solver simulators. Furthermore, we compare the results of a classical implementation with the performance obtained using a quantum simulator.

Speaker: Laura Cappelli (Istituto Nazionale di Fisica Nucleare)

QuantumCreditScoring2.pdf

09:40 Quantum image processing and optical neural networks

Visual information can be manipulated in terms of images, usually captured and then processed through a sequence of computational operations. Alternatively, optical systems can perform such operations directly, reducing computational overhead at the cost of stricter design requirements. We discuss this workflow in the context of quantum technologies. First, we introduce a quantum computational algorithm that uses the quantum Fourier transform to discard the high spatial-frequency qubits of an image, downsampling it to a lower resolution. Our method allows us to capture, compress, and communicate visual information (or high-dimensional natural data) even with limited resources [1,2]. Then, we present a quantum optical pattern recognition method for binary classification tasks. Leveraging the Hong-Ou-Mandel effect, this setup is specifically designed to reproduce classical perceptrons and shallow neural networks. Our method classifies patterns without reconstructing their images, encoding the spatial information of the object in the spectrum of a single photon, providing a superexponential speedup over classical methods [3,4].

References
[1] Simone Roncallo, Lorenzo Maccone and Chiara Macchiavello, Quantum JPEG, AVS Quantum Sci. 5, 043803 (2023)
[2] Emanuele Tumbiolo, Simone Roncallo, Chiara Macchiavello and Lorenzo Maccone, Quantum frequency resampling, npj Quantum Inf. 11, 123 (2025)
[3] Simone Roncallo, Angela Rosy Morgillo, Chiara Macchiavello, Lorenzo Maccone and Seth Lloyd, Quantum optical classifier with superexponential speedup, Commun. Phys. 8 147 (2025)
[4] Simone Roncallo, Angela Rosy Morgillo, Seth Lloyd, Chiara Macchiavello and Lorenzo Maccone, Quantum optical shallow networks, arXiv.2507.21036

Speaker: Dr Simone Roncallo (University of Pavia)

INFNWorkshopMilano2026.pdf

10:00 → 10:45 Technological aspects

Dedicated to hardware platforms, device technologies, and implementation issues in quantum information science.

Convener: Dr Andrea Giachero (INFN Milano-Bicocca)

10:00 Two-photon interferometry: beyond the boundaries of distinguishability

Two-photon interferometry (TPI), the very investigation of which historically led to the development of the modern description of the coherence of light as well as to the foundation of quantum optics [1,2,3], has evolved towards a booming research field whose applications span from quantum cryptography and computing [4] to biosensing [5], astrometry and telescopy [6], imaging and tomography [7,8], metrology and fundamental physics [9]. Photon “bunching”, a direct consequence of the bosonic nature of light and one of the main features of TPI, is the working principle behind the production of optical N00N states, candidates of strong interest for quantum communications and computing [10] as well as for the collective technologies that are now referred to as “quantum internet”. Intriguingly, TPI can be investigated with many different probe states, from entangled pairs on two or more degrees of freedom to coherent and thermal light, proving its remarkable capability to adapt to a wide variety of sources and needs. In this scientific and technological landscape, even the intrinsic boundary imposed to the TPI operating range by the requirement of total or at least partial indistinguishability of the optical paths has ultimately come under scrutiny [11,12]. In this talk, an innovative approach to TPI will be presented [13,14], showcasing how conjugate-variable interrogation of a TPI system leads to the emergence of a rich dynamics characterized by broader operative range and multiplexing capabilities, both highly desirable qualities for communication and cryptography protocols.

Speaker: Dr Fabrizio Sgobba (Università degli studi di Bari "Aldo Moro", Istituto Nazionale di Fisica Nucleare)

Workshop Quantum Computing INFN Bicocca-Sgobba.pptx

10:20 Integrated Waveguide Source of Squeezed Vacuum for Quantum computation and Gravitational-Wave Detection

Squeezed vacuum states are a key quantum resource for continuous variable quantum computing, quantum communications, and quantum metrology since they enable measurements below the shot-noise limit. Although state-of-the-art squeezed-light sources typically rely on bulk optical parametric oscillators [1], integrated nonlinear waveguides provide a promising alternative capable of reducing system complexity, enhancing stability, and supporting scalable deployment in quantum-enhanced sensing platforms.

In this work, we demonstrate the generation of a squeezed vacuum state in a periodically poled waveguide made on a lithium niobate substrate and implement feedback control [2] to stabilize the squeezing within the frequency band relevant for quantum computation (MHz-GHz) [3] and gravitational-wave detection (10 Hz - 10 kHz). Our objective is to demonstrate the feasibility of combining integrated squeezed-light sources with active control techniques, paving the way for compact and deployable quantum-enhancement modules for next-generation computing machines and precision metrology systems [4].

Speaker: Hamza Hasnaoui (Istituto Nazionale di Fisica Nucleare)

Integrated squeezing source.pptx

10:45 → 11:15 Coffee / tea break

11:15 → 13:00 Quantum Simulation

Digital and analog simulation of physical systems, including simulation
methods, algorithms, and practical use-cases.

Convener: Concezio Bozzi (Istituto Nazionale di Fisica Nucleare)

11:15 Modelling and Simulation of Dispersively Coupled Resonator–Qubit Systems under Quantum and Semiclassical Drives

We study a class of driven dispersive resonator–qubit systems, consisting of a single resonator coupled to a network of qubits, with the interaction treated in the dispersive regime. The system dynamics is analyzed under different types of input fields, ranging from semiclassical electromagnetic drives to single-photon quantum excitations.

We develop a set of simplified theoretical models that capture the essential features of dynamics while retaining partial analytical tractability. These models are complemented by numerical simulations based on open quantum system techniques, implemented using the QuTiP framework. The resulting framework constitutes a flexible toolbox for the systematic study of dispersive cavity–qubit architectures, applicable to single-qubit and multi-qubit configurations and largely independent of the specific physical realization of the qubits. The methods can be extended and integrated with optimization and machine-learning techniques, enabling the efficient exploration of complex parameter spaces and protocol design.

While the approach is general and applicable to different qubit platforms, superconducting qubits provide a natural testbed for validating the proposed models. Possible applications include the analysis and optimization of driven quantum systems and their use in quantum simulation and information-processing tasks, and in metrology and sensing.

Speaker: Fabio Chiarello (IFN-CNR)

11:40 Maximum Independent Set via Probabilistic and Quantum Cellular Automata

We study probabilistic cellular automata (PCA) and quantum cellular automata (QCA) as frameworks for solving the Maximum Independent Set (MIS) problem. We first introduce a synchronous PCA whose dynamics drives
the system toward the manifold of maximal independent sets. Numerical evidence shows that the MIS convergence
probability increases significantly as the activation probability p → 1, and we characterize how the steps required to
reach the absorbing state scale with system size and graph connectivity. Motivated by this behavior, we construct
a QCA combining a pure dissipative phase with a constraint-preserving unitary evolution that redistributes probability within this manifold. Tensor Network simulations reveal that repeated dissipative–unitary cycles concentrate
population on MIS configurations. We also provide an empirical estimate of how the convergence time scales with
graph size, suggesting that QCA dynamics can provide an efficient alternative to adiabatic and variational quantum
optimization methods based exclusively on local and translationally invariant rules.

Speaker: Matteo Grotti (Istituto Nazionale di Fisica Nucleare)

QCA_MIS.pdf

QCA_MIS.pptx

12:05 Improving VQE Ground State Accuracy via Soft-Coded Orthogonal Subspace Representations

In this talk, we present a new approach to improve the accuracy of ground state approximations in Variational Quantum Eigensolver (VQE) algorithms. We employ subspace representations where orthogonality is enforced via "soft-coded" constraints within the cost function, rather than "hard-coded" at the circuit level.

Similar to other subspace-based methods like Subspace-Search VQE (SSVQE) and Multistate Contracted VQE (MCVQE), our method trains parameters to overlap with the low-energy sector before diagonalizing the Hamiltonian restricted to the subspace. However, by shifting the orthogonality constraints via penalty terms during the minimization, we find that significantly shallower quantum circuits can be used while maintaining high fidelity.

We validate this approach on two benchmark cases: a 3x3 transverse-field Ising model and random realizations of the Edwards-Anderson spin-glass model on a 4x4 lattice. We show that our soft-coded representation outperforms single-state (standard VQE) and multi-state (SSVQE/MCVQE) approaches, offering a favorable trade-off between circuit depth and accuracy.

Speaker: Marco Intini (Università di Pisa, INFN Pisa)

Intini_v1.pdf