arXiv subjects

Massimo Panella

Publications and source records attributed to Massimo Panella.

At least 19 recordsLinked to original sources

arXiv2026

Gated QKAN-FWP: Scalable Quantum-inspired Sequence Learning

Fast Weight Programmers (FWPs) encode temporal dependencies through dynamically updated parameters rather than recurrent hidden states. Quantum FWPs (QFWPs) extend this idea with variational quantum circuits (VQCs), but existing implementations rely on multi-qubit architectures that are difficult to scale on noisy intermediate-scale quantum (NISQ) devices and expensive to simulate classically. We propose gated QKAN-FWP, a fast-weight framework that integrates FWP with Quantum-inspired Kolmogorov-Arnold Network (QKAN) using single-qubit data re-uploading circuits as learnable nonlinear activation, known as DatA Re-Uploading ActivatioN (DARUAN). We further introduce a scalar-gated fast-weight update rule that stabilizes parameter evolution, supported by a theoretical analysis of its adaptive memory kernel, geometric boundedness, and parallelizable gradient paths. We evaluate the framework across time-series benchmarks, MiniGrid reinforcement learning, and highlight real-world solar cycle forecasting as our main practical result. In the long-horizon setting with 528-month input window and 132-month forecast horizon, our 12.5k-parameter model achieves lower scaled Mean Square Error (MSE), peak amplitude error, and peak timing error than a suite of classical recurrent baselines with up to 13x more parameters, including Long Short-Term Memory (LSTM) networks (25.9k-89.1k parameters), WaveNet-LSTM (167k), Vanilla recurrent neural network (11.5k), and a Modified Echo State Network (132k). To validate NISQ compatibility, we further deploy the trained fast programmer on IonQ and IBM Quantum processors, recovering forecasting accuracy within 0.1% relative MSE of the noiseless simulator at 1024 shots. These results position gated QKAN-FWP as a scalable, parameter-efficient, and NISQ-compatible approach to quantum-inspired sequence modeling.

cs.LG↗

arXiv2025

Enriching Earth Observation labeled data with Quantum Conditioned Diffusion Models

The rapid adoption of diffusion models (DMs) in the Earth Observation (EO) domain has unlocked new generative capabilities aimed at producing new samples, whose statistical properties closely match real imagery, for tasks such as synthesizing missing data, augmenting scarce labeled datasets, and improving image reconstruction. This is particularly relevant in EO, where labeled data are often costly to obtain and limited in availability. However, classical DMs still face significant computational limitations, requiring hundreds to thousands of inference steps, as well as difficulties in capturing the intricate spatial and spectral correlations characteristic of EO data. Recent research in Quantum Machine Learning (QML), including initial attempts of Quantum Generative Models, offers a fundamentally different approach to overcome these challenges. Motivated by these considerations, we introduce the Quanvolutional Conditioned U-Net (QCU-Net), a hybrid quantum--classical architecture that applies quantum operations within a conditioned diffusion framework using a novel quanvolutional feature-extraction approach, for generating synthetic labeled EO imagery. Extensive experiments on the EuroSAT RGB dataset demonstrate that our QCU-Net achieves superior results. Notably, it reduces the Fr\'echet Inception Distance by 64%, lowers the Kernel Inception Distance by 76%, and yields higher semantic accuracy. Ablation studies further reveal that strategically positioning quantum layers and employing entangling variational circuits enhance model performance and convergence. This work represents the first successful adaptation of class-conditioned quantum diffusion modeling in the EO domain, paving the way for quantum-enhanced remote sensing imagery synthesis.

quant-ph↗

arXiv2025

Quantum Latent Diffusion Models

The introduction of quantum concepts is increasingly making its way into generative machine learning models. However, while there are various implementations of quantum Generative Adversarial Networks, the integration of quantum elements into diffusion models remains an open and challenging task. In this work, we propose a potential version of a quantum diffusion model that leverages the established idea of classical latent diffusion models. This involves using a traditional autoencoder to reduce images, followed by operations with variational circuits in the latent space. To effectively assess the benefits brought by quantum computing, the images generated by the quantum latent diffusion model have been compared to those generated by a classical model with a similar number of parameters, evaluated in terms of quantitative metrics. The results demonstrate an advantage in using a quantum version, as evidenced by obtaining better metrics for the images generated by the quantum version compared to those obtained by the classical version. Furthermore, quantum models continue to outperform even when considering small percentages of the dataset for training, demonstrating the quantum's ability to extract features more effectively even in a few shot learning scenario.

quant-ph↗

arXiv2025

Hyperdimensional Computing for ADHD Classification using EEG Signals

Following the recent interest in applying the Hyperdimensional Computing paradigm in medical context to power up the performance of general machine learning applied to biomedical data, this study represents the first attempt at employing such techniques to solve the problem of classification of Attention Deficit Hyperactivity Disorder using electroencephalogram signals. Making use of a spatio-temporal encoder, and leveraging the properties of HDC, the proposed model achieves an accuracy of 88.9%, outperforming traditional Deep Neural Networks benchmark models. The core of this research is not only to enhance the classification accuracy of the model but also to explore its efficiency in terms of the required training data: a critical finding of the study is the identification of the minimum number of patients needed in the training set to achieve a sufficient level of accuracy. To this end, the accuracy of our model trained with only $7$ of the $79$ patients is comparable to the one from benchmarks trained on the full dataset. This finding underscores the model's efficiency and its potential for quick and precise ADHD diagnosis in medical settings where large datasets are typically unattainable.

eess.SP↗

arXiv2025

Quantum Simplicial Neural Networks

Graph Neural Networks (GNNs) excel at learning from graph-structured data but are limited to modeling pairwise interactions, insufficient for capturing higher-order relationships present in many real-world systems. Topological Deep Learning (TDL) has allowed for systematic modeling of hierarchical higher-order interactions by relying on combinatorial topological spaces such as simplicial complexes. In parallel, Quantum Neural Networks (QNNs) have been introduced to leverage quantum mechanics for enhanced computational and learning power. In this work, we present the first Quantum Topological Deep Learning Model: Quantum Simplicial Networks (QSNs), being QNNs operating on simplicial complexes. QSNs are a stack of Quantum Simplicial Layers, which are inspired by the Ising model to encode higher-order structures into quantum states. Experiments on synthetic classification tasks show that QSNs can outperform classical simplicial TDL models in accuracy and efficiency, demonstrating the potential of combining quantum computing with TDL for processing data on combinatorial topological spaces.

cs.NE↗

arXiv2024

Q-SCALE: Quantum computing-based Sensor Calibration for Advanced Learning and Efficiency

In a world burdened by air pollution, the integration of state-of-the-art sensor calibration techniques utilizing Quantum Computing (QC) and Machine Learning (ML) holds promise for enhancing the accuracy and efficiency of air quality monitoring systems in smart cities. This article investigates the process of calibrating inexpensive optical fine-dust sensors through advanced methodologies such as Deep Learning (DL) and Quantum Machine Learning (QML). The objective of the project is to compare four sophisticated algorithms from both the classical and quantum realms to discern their disparities and explore possible alternative approaches to improve the precision and dependability of particulate matter measurements in urban air quality surveillance. Classical Feed-Forward Neural Networks (FFNN) and Long Short-Term Memory (LSTM) models are evaluated against their quantum counterparts: Variational Quantum Regressors (VQR) and Quantum LSTM (QLSTM) circuits. Through meticulous testing, including hyperparameter optimization and cross-validation, the study assesses the potential of quantum models to refine calibration performance. Our analysis shows that: the FFNN model achieved superior calibration accuracy on the test set compared to the VQR model in terms of lower L1 loss function (2.92 vs 4.81); the QLSTM slightly outperformed the LSTM model (loss on the test set: 2.70 vs 2.77), despite using fewer trainable weights (66 vs 482).

cs.LG↗

arXiv2024

Guess What I Think: Streamlined EEG-to-Image Generation with Latent Diffusion Models

Generating images from brain waves is gaining increasing attention due to its potential to advance brain-computer interface (BCI) systems by understanding how brain signals encode visual cues. Most of the literature has focused on fMRI-to-Image tasks as fMRI is characterized by high spatial resolution. However, fMRI is an expensive neuroimaging modality and does not allow for real-time BCI. On the other hand, electroencephalography (EEG) is a low-cost, non-invasive, and portable neuroimaging technique, making it an attractive option for future real-time applications. Nevertheless, EEG presents inherent challenges due to its low spatial resolution and susceptibility to noise and artifacts, which makes generating images from EEG more difficult. In this paper, we address these problems with a streamlined framework based on the ControlNet adapter for conditioning a latent diffusion model (LDM) through EEG signals. We conduct experiments and ablation studies on popular benchmarks to demonstrate that the proposed method beats other state-of-the-art models. Unlike these methods, which often require extensive preprocessing, pretraining, different losses, and captioning models, our approach is efficient and straightforward, requiring only minimal preprocessing and a few components. The code is available at https://github.com/LuigiSigillo/GWIT.

cs.CV↗

arXiv2024

On the Effects of Small Graph Perturbations in the MaxCut Problem by QAOA

We investigate the Maximum Cut (MaxCut) problem on different graph classes with the Quantum Approximate Optimization Algorithm (QAOA) using symmetries. In particular, heuristics on the relationship between graph symmetries and the approximation ratio achieved by a QAOA simulation are considered. To do so, we first solve the MaxCut problem on well-known graphs, then we consider a simple and controllable perturbation of the graph and find again the approximate MaxCut with the QAOA. Through an analysis of the spectrum of the graphs and their perturbations, as well as a careful study of the associated automorphism groups, we aim to extract valuable insights into how symmetry impacts the performance of QAOA. These insights can then be leveraged to heuristically reduce the quantum circuit complexity, the number of training steps, or the number of parameters involved, thus enhancing the efficiency and effectiveness of QAOA-based solutions.

quant-ph↗

arXiv2024

From Graphs to Qubits: A Critical Review of Quantum Graph Neural Networks

Quantum Graph Neural Networks (QGNNs) represent a novel fusion of quantum computing and Graph Neural Networks (GNNs), aimed at overcoming the computational and scalability challenges inherent in classical GNNs that are powerful tools for analyzing data with complex relational structures but suffer from limitations such as high computational complexity and over-smoothing in large-scale applications. Quantum computing, leveraging principles like superposition and entanglement, offers a pathway to enhanced computational capabilities. This paper critically reviews the state-of-the-art in QGNNs, exploring various architectures. We discuss their applications across diverse fields such as high-energy physics, molecular chemistry, finance and earth sciences, highlighting the potential for quantum advantage. Additionally, we address the significant challenges faced by QGNNs, including noise, decoherence, and scalability issues, proposing potential strategies to mitigate these problems. This comprehensive review aims to provide a foundational understanding of QGNNs, fostering further research and development in this promising interdisciplinary field.

quant-ph↗

arXiv2024

A Study on Quantum Graph Neural Networks Applied to Molecular Physics

This paper introduces a novel architecture for Quantum Graph Neural Networks, which is significantly different from previous approaches found in the literature. The proposed approach produces similar outcomes with respect to previous models but with fewer parameters, resulting in an extremely interpretable architecture rooted in the underlying physics of the problem. The architectural novelties arise from three pivotal aspects. Firstly, we employ an embedding updating method that is analogous to classical Graph Neural Networks, therefore bridging the classical-quantum gap. Secondly, each layer is devoted to capturing interactions of distinct orders, aligning with the physical properties of the system. Lastly, we harness SWAP gates to emulate the problem's inherent symmetry, a novel strategy not found currently in the literature. The obtained results in the considered experiments are encouraging to lay the foundation for continued research in this field.

quant-ph↗

arXiv2024

An Explainable Fast Deep Neural Network for Emotion Recognition

In the context of artificial intelligence, the inherent human attribute of engaging in logical reasoning to facilitate decision-making is mirrored by the concept of explainability, which pertains to the ability of a model to provide a clear and interpretable account of how it arrived at a particular outcome. This study explores explainability techniques for binary deep neural architectures in the framework of emotion classification through video analysis. We investigate the optimization of input features to binary classifiers for emotion recognition, with face landmarks detection using an improved version of the Integrated Gradients explainability method. The main contribution of this paper consists in the employment of an innovative explainable artificial intelligence algorithm to understand the crucial facial landmarks movements during emotional feeling, using this information also for improving the performances of deep learning-based emotion classifiers. By means of explainability, we can optimize the number and the position of the facial landmarks used as input features for facial emotion recognition, lowering the impact of noisy landmarks and thus increasing the accuracy of the developed models. In order to test the effectiveness of the proposed approach, we considered a set of deep binary models for emotion classification trained initially with a complete set of facial landmarks, which are progressively reduced based on a suitable optimization procedure. The obtained results prove the robustness of the proposed explainable approach in terms of understanding the relevance of the different facial points for the different emotions, also improving the classification accuracy and diminishing the computational cost.

cs.CV↗

arXiv2024

Towards Efficient Quantum Hybrid Diffusion Models

In this paper, we propose a new methodology to design quantum hybrid diffusion models, derived from classical U-Nets with ResNet and Attention layers. Specifically, we propose two possible different hybridization schemes combining quantum computing's superior generalization with classical networks' modularity. In the first one, we acted at the vertex: ResNet convolutional layers are gradually replaced with variational circuits to create Quantum ResNet blocks. In the second proposed architecture, we extend the hybridization to the intermediate level of the encoder, due to its higher sensitivity in the feature extraction process. In order to conduct an in-depth analysis of the potential advantages stemming from the integration of quantum layers, images generated by quantum hybrid diffusion models are compared to those generated by classical models, and evaluated in terms of several quantitative metrics. The results demonstrate an advantage in using a hybrid quantum diffusion models, as they generally synthesize better-quality images and converges faster. Moreover, they show the additional advantage of having a lower number of parameters to train compared to the classical one, with a reduction that depends on the extent to which the vertex is hybridized.

quant-ph↗

arXiv2023

A General Approach to Dropout in Quantum Neural Networks

In classical Machine Learning, "overfitting" is the phenomenon occurring when a given model learns the training data excessively well, and it thus performs poorly on unseen data. A commonly employed technique in Machine Learning is the so called "dropout", which prevents computational units from becoming too specialized, hence reducing the risk of overfitting. With the advent of Quantum Neural Networks as learning models, overfitting might soon become an issue, owing to the increasing depth of quantum circuits as well as multiple embedding of classical features, which are employed to give the computational nonlinearity. Here we present a generalized approach to apply the dropout technique in Quantum Neural Network models, defining and analysing different quantum dropout strategies to avoid overfitting and achieve a high level of generalization. Our study allows to envision the power of quantum dropout in enabling generalization, providing useful guidelines on determining the maximal dropout probability for a given model, based on overparametrization theory. It also highlights how quantum dropout does not impact the features of the Quantum Neural Networks model, such as expressibility and entanglement. All these conclusions are supported by extensive numerical simulations, and may pave the way to efficiently employing deep Quantum Machine Learning models based on state-of-the-art Quantum Neural Networks.

quant-ph↗

arXiv2023

Resource Saving via Ensemble Techniques for Quantum Neural Networks

Quantum neural networks hold significant promise for numerous applications, particularly as they can be executed on the current generation of quantum hardware. However, due to limited qubits or hardware noise, conducting large-scale experiments often requires significant resources. Moreover, the output of the model is susceptible to corruption by quantum hardware noise. To address this issue, we propose the use of ensemble techniques, which involve constructing a single machine learning model based on multiple instances of quantum neural networks. In particular, we implement bagging and AdaBoost techniques, with different data loading configurations, and evaluate their performance on both synthetic and real-world classification and regression tasks. To assess the potential performance improvement under different environments, we conduct experiments on both simulated, noiseless software and IBM superconducting-based QPUs, suggesting these techniques can mitigate the quantum hardware noise. Additionally, we quantify the amount of resources saved using these ensemble techniques. Our findings indicate that these methods enable the construction of large, powerful models even on relatively small quantum devices.

quant-ph↗

arXiv2022

Quasi-Chaotic Oscillators Based on Modular Quantum Circuits

Digital circuits based on residue number systems have been considered to produce a pseudo-random behavior. The present work is an initial step towards the complete implementation of those systems for similar applications using quantum technology. We propose the implementation of a quasi-chaotic oscillator based on quantum modular addition and multiplication and we prove that quantum computing allows the parallel processing of data, paving the way for a fast and robust multi-channel encryption/decryption scheme. The resulting structure is assessed by several experiments in order to ascertain the desired noise-like behavior.

quant-ph↗

arXiv2021

On Effects of Compression with Hyperdimensional Computing in Distributed Randomized Neural Networks

A change of the prevalent supervised learning techniques is foreseeable in the near future: from the complex, computational expensive algorithms to more flexible and elementary training ones. The strong revitalization of randomized algorithms can be framed in this prospect steering. We recently proposed a model for distributed classification based on randomized neural networks and hyperdimensional computing, which takes into account cost of information exchange between agents using compression. The use of compression is important as it addresses the issues related to the communication bottleneck, however, the original approach is rigid in the way the compression is used. Therefore, in this work, we propose a more flexible approach to compression and compare it to conventional compression algorithms, dimensionality reduction, and quantization techniques.

cs.LG↗

arXiv2021

Hyperdimensional Computing for Efficient Distributed Classification with Randomized Neural Networks

In the supervised learning domain, considering the recent prevalence of algorithms with high computational cost, the attention is steering towards simpler, lighter, and less computationally extensive training and inference approaches. In particular, randomized algorithms are currently having a resurgence, given their generalized elementary approach. By using randomized neural networks, we study distributed classification, which can be employed in situations were data cannot be stored at a central location nor shared. We propose a more efficient solution for distributed classification by making use of a lossy compression approach applied when sharing the local classifiers with other agents. This approach originates from the framework of hyperdimensional computing, and is adapted herein. The results of experiments on a collection of datasets demonstrate that the proposed approach has usually higher accuracy than local classifiers and getting close to the benchmark - the centralized classifier. This work can be considered as the first step towards analyzing the variegated horizon of distributed randomized neural networks.

cs.LG↗

arXiv2020

Perceptron Theory Can Predict the Accuracy of Neural Networks

Multilayer neural networks set the current state of the art for many technical classification problems. But, these networks are still, essentially, black boxes in terms of analyzing them and predicting their performance. Here, we develop a statistical theory for the one-layer perceptron and show that it can predict performances of a surprisingly large variety of neural networks with different architectures. A general theory of classification with perceptrons is developed by generalizing an existing theory for analyzing reservoir computing models and connectionist models for symbolic reasoning known as vector symbolic architectures. Our statistical theory offers three formulas leveraging the signal statistics with increasing detail. The formulas are analytically intractable, but can be evaluated numerically. The description level that captures maximum details requires stochastic sampling methods. Depending on the network model, the simpler formulas already yield high prediction accuracy. The quality of the theory predictions is assessed in three experimental settings, a memorization task for echo state networks (ESNs) from reservoir computing literature, a collection of classification datasets for shallow randomly connected networks, and the ImageNet dataset for deep convolutional neural networks. We find that the second description level of the perceptron theory can predict the performance of types of ESNs, which could not be described previously. The theory can predict deep multilayer neural networks by being applied to their output layer. While other methods for prediction of neural networks performance commonly require to train an estimator model, the proposed theory requires only the first two moments of the distribution of the postsynaptic sums in the output neurons. The perceptron theory compares favorably to other methods that do not rely on training an estimator model.

cs.LG↗