Marco Romanelli

35 papers A* 6A 4B 1Journal 20Unranked 3
YearRankTypeTitle / Venue / Authors
2025 J jnl
Happiness as a Measure of Fairness.
CoRR
Georg Pichler, Marco Romanelli, Pablo Piantanida
2025 A conf
Optimal Zero-shot Regret Minimization for Selective Classification with Out-of-Distribution Detection.
UAI
Eduardo Dadalto Câmara Gomes, Marco Romanelli
2025 J jnl
The Disparate Impacts of Speculative Decoding.
CoRR
Jameson Sandler, Ahmet Üstün, Marco Romanelli, Sara Hooker, Ferdinando Fioretto
2024 A* conf
A Data-Driven Measure of Relative Uncertainty for Misclassification Detection.
ICLR
Eduardo Dadalto Câmara Gomes, Marco Romanelli, Georg Pichler, Pablo Piantanida
2024 A* conf
Beyond the Norms: Detecting Prediction Errors in Regression Models.
ICML
Andrés Altieri, Marco Romanelli, Georg Pichler, Florence Alberge, Pablo Piantanida
2024 J jnl
Beyond the Norms: Detecting Prediction Errors in Regression Models.
CoRR
Andrés Altieri, Marco Romanelli, Georg Pichler, Florence Alberge, Pablo Piantanida
2024 A* conf
Disparate Impact on Group Accuracy of Linearization for Private Inference.
ICML
Saswat Das, Marco Romanelli, Ferdinando Fioretto
2024 J jnl
Disparate Impact on Group Accuracy of Linearization for Private Inference.
CoRR
Saswat Das, Marco Romanelli, Ferdinando Fioretto
2024 J jnl
Low-rank finetuning for LLMs: A fairness perspective.
CoRR
Saswat Das, Marco Romanelli, Cuong Tran, Zarreen Reza, Bhavya Kailkhura, Ferdinando Fioretto
2024 A conf
On the (In)feasibility of ML Backdoor Detection as an Hypothesis Testing Problem.
AISTATS
Georg Pichler, Marco Romanelli, Divya Prakash Manivannan, Prashanth Krishnamurthy, Farshad Khorrami, Siddharth Garg
2024 J jnl
On the (In)feasibility of ML Backdoor Detection as an Hypothesis Testing Problem.
CoRR
Georg Pichler, Marco Romanelli, Divya Prakash Manivannan, Prashanth Krishnamurthy, Farshad Khorrami, Siddharth Garg
2024 A conf
Optimal Zero-Shot Detector for Multi-Armed Attacks.
AISTATS
Federica Granese, Marco Romanelli, Pablo Piantanida
2024 J jnl
Optimal Zero-Shot Detector for Multi-Armed Attacks.
CoRR
Federica Granese, Marco Romanelli, Pablo Piantanida
2024 J jnl
Perfectly Accurate Membership Inference by a Dishonest Central Server in Federated Learning.
IEEE Trans. Dependable Secur. Comput.
Georg Pichler, Marco Romanelli, Leonardo Rey Vega, Pablo Piantanida
2024 A* conf
Retrieval-Guided Reinforcement Learning for Boolean Circuit Minimization.
ICLR
Animesh Basak Chowdhury, Marco Romanelli, Benjamin Tan, Ramesh Karri, Siddharth Garg
2024 J jnl
Retrieval-Guided Reinforcement Learning for Boolean Circuit Minimization.
CoRR
Animesh Basak Chowdhury, Marco Romanelli, Benjamin Tan, Ramesh Karri, Siddharth Garg
2023 J jnl
A Data-Driven Measure of Relative Uncertainty for Misclassification Detection.
CoRR
Eduardo Dadalto, Marco Romanelli, Georg Pichler, Pablo Piantanida
2023 J jnl
A Halfspace-Mass Depth-Based Method for Adversarial Attack Detection.
Trans. Mach. Learn. Res.
Marine Picot, Federica Granese, Guillaume Staerman, Marco Romanelli, Francisco Messina, Pablo Piantanida, Pierre Colombo
2023 J jnl
A Minimax Approach Against Multi-Armed Adversarial Attacks Detection.
CoRR
Federica Granese, Marco Romanelli, Siddharth Garg, Pablo Piantanida
2023 J jnl
INVICTUS: Optimizing Boolean Logic Circuit Synthesis via Synergistic Learning and Search.
CoRR
Animesh Basak Chowdhury, Marco Romanelli, Benjamin Tan, Ramesh Karri, Siddharth Garg
2022 — conf
MEAD: A Multi-Armed Approach for Evaluation of Adversarial Examples Detectors.
ECML/PKDD (3)
Federica Granese, Marine Picot, Marco Romanelli, Francesco Messina, Pablo Piantanida
2022 J jnl
MEAD: A Multi-Armed Approach for Evaluation of Adversarial Examples Detectors.
CoRR
Federica Granese, Marine Picot, Marco Romanelli, Francisco Messina, Pablo Piantanida
2022 J jnl
Perfectly Accurate Membership Inference by a Dishonest Central Server in Federated Learning.
CoRR
Georg Pichler, Marco Romanelli, Leonardo Rey Vega, Pablo Piantanida
2021 A* conf
DOCTOR: A Simple Method for Detecting Misclassification Errors.
NeurIPS
Federica Granese, Marco Romanelli, Daniele Gorla, Catuscia Palamidessi, Pablo Piantanida
2021 J jnl
DOCTOR: A Simple Method for Detecting Misclassification Errors.
CoRR
Federica Granese, Marco Romanelli, Daniele Gorla, Catuscia Palamidessi, Pablo Piantanida
2020 — conf
Derivation of Constraints from Machine Learning Models and Applications to Security and Privacy.
Gabbrielli's Festschrift
Moreno Falaschi, Catuscia Palamidessi, Marco Romanelli
2020 A* conf
Estimating g-Leakage via Machine Learning.
CCS
Marco Romanelli, Konstantinos Chatzikokolakis, Catuscia Palamidessi, Pablo Piantanida
2020 J jnl
Estimating g-Leakage via Machine Learning.
CoRR
Marco Romanelli, Konstantinos Chatzikokolakis, Catuscia Palamidessi, Pablo Piantanida
2020 J jnl
Feature selection in machine learning: Rényi min-entropy vs Shannon entropy.
CoRR
Catuscia Palamidessi, Marco Romanelli
2020 — —
Machine learning methods for privacy protection : leakage measurement and mechanisms design. (Méthodes d'apprentissage machine pour la protection de la vie privée : mesure de leakage et design des mécanismes).
Marco Romanelli
2020 B conf
Modern Applications of Game-Theoretic Principles (Invited Paper).
CONCUR
Catuscia Palamidessi, Marco Romanelli
2020 A conf
Optimal Obfuscation Mechanisms via Machine Learning.
CSF
Marco Romanelli, Konstantinos Chatzikokolakis, Catuscia Palamidessi
2019 J jnl
Generating Optimal Privacy-Protection Mechanisms via Machine Learning.
CoRR
Marco Romanelli, Catuscia Palamidessi, Konstantinos Chatzikokolakis
2019 J jnl
STRAIN: an R package for multi-locus sequence typing from whole genome sequencing data.
BMC Bioinform.
Mattia Dalsass, Margherita Bodini, Christophe G. Lambert, Marie-Cécile Mortier, Marco Romanelli, Duccio Medini, Alessandro Muzzi, Alessandro Brozzi
2018 — conf
Feature Selection with Rényi Min-Entropy.
ANNPR
Catuscia Palamidessi, Marco Romanelli