Selected publications
Selected publications
Selected publications in reservoir computing, time-series machine learning, anomaly detection, generative modelling and theoretical mathematics.
Under review
Free-Probability Kernels for Zero-Rollout Hyperparameter Selection in Reservoir Computing
Sara Malacarne, Andrea Ceni, Claudio Gallicchio.
Submitted to IEEE Transactions on Neural Networks and Learning Systems, 2026.
Scalable Context-Aware Graph Attention for Unsupervised Anomaly Detection in Large-Scale Mobile Networks
Eirik Hoel-Hoiseth, Erlend Aune, David Zsolt Biro, Massimiliano Ruocco.
Submitted to IEEE Transactions on Network and Service Management, 2026.
Published
Context-Aware Graph Attention for Unsupervised Telco Anomaly Detection
Sara Malacarne, Eirik Hoel-Hoiseth, Erlend Aune, David Zsolt Biro, Massimiliano Ruocco.
ESANN, 2026.
Develops an unsupervised graph-attention model for anomaly detection in multivariate mobile-network time series, using context embeddings and label-free threshold calibration.
Closing the Gap Between Synthetic and Ground Truth Time Series Distributions via Neural Mapping
Daesoo Lee, Sara Malacarne, Erlend Aune.
Workshop on Machine Learning for Irregular Time Series (ML4ITS), ECML PKDD, 2025.
Introduces neural mapping methods to improve the fidelity of synthetic time-series distributions.
Explainable Time Series Anomaly Detection Using Masked Latent Generative Modeling
Daesoo Lee, Sara Malacarne, Erlend Aune.
Pattern Recognition, 156, 2024.
Uses masked latent generative modelling to support explainable anomaly detection in time-series data.
Circle Attention: Forecasting Network Traffic by Learning Interpretable Spatial Relationships from Intersecting Circles
Espen Haugsdal, Sara Malacarne, Massimiliano Ruocco.
ECML PKDD, pp. 106-121, 2023.
Introduces an interpretable spatial attention mechanism for forecasting telecommunication network traffic.
Vector Quantized Time Series Generation with a Bidirectional Prior Model
Daesoo Lee, Sara Malacarne, Erlend Aune.
AISTATS, Proceedings of Machine Learning Research, pp. 7665-7693, 2023.
Introduces TimeVQVAE, a vector-quantized generative model for time-series data with bidirectional transformer priors.
AI Anomaly Detection for Cloudified Mobile Core Architectures
Foivos Michelinakis, Joan S. Pujol-Roig, Sara Malacarne, Min Xie, Thomas Dreibholz, Sayantini Majumdar, Wint Yi Poe, Georgios Patounas, Maria Carmen Guerrero Lopez, Ahmed Elmokashfi, Vasileios Theodorou.
IEEE Transactions on Network and Service Management, 20(2):1976-1992, 2022.
Applies deep-learning anomaly detection to cloudified mobile-core architectures for service assurance and root-cause analysis.
An Exposed Closed-Loop Model for Customer-Driven Service Assurance Automation
Min Xie, Foivos Michelinakis, Thomas Dreibholz, Joan S. Pujol-Roig, Sara Malacarne, Sayantini Majumdar, Wint Yi Poe, Ahmed Elmokashfi.
Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit), pp. 419-424, 2021.
Proposes a closed-loop model for customer-driven service assurance automation in networked systems.
Martin Boundaries of the Duals of Free Unitary Quantum Groups
Sara Malacarne, Sergey Neshveyev.
Compositio Mathematica, 155(6):1171-1193, 2019.
Computes Martin boundaries for duals of free unitary quantum groups, contributing to the probabilistic boundary theory of quantum groups.
Woronowicz Tannaka-Krein Duality and Free Orthogonal Quantum Groups
Sara Malacarne.
Mathematica Scandinavica, 122(1):151-160, 2018.
Gives a short proof of Woronowicz Tannaka-Krein duality and relates it to free orthogonal quantum groups.
Probabilistic Boundaries of Finite Extensions of Quantum Groups
Sara Malacarne, Sergey Neshveyev.
Infinite Dimensional Analysis, Quantum Probability and Related Topics, 20(4), 2017.
Studies Poisson and Martin boundaries for finite extensions of discrete quantum groups.
Complete publication record
A complete and updated publication list is available on Google Scholar.