Mario Díaz

43 papers A 1B 14Journal 28
YearRankTypeTitle / Venue / Authors
2026 J jnl
Optimality of Staircase Mechanisms for Vector Queries under Differential Privacy.
CoRR
James Melbourne, Mario Díaz, Shahab Asoodeh
2025 B conf
Auditing Privacy of Additive Noise Mechanisms Using Linear Predictive Models.
ISIT
Monica Welfert, Nathan Stromberg, Mario Díaz, James Melbourne, Lalitha Sankar
2025 A conf
Locally Private Sampling with Public Data.
AISTATS
Behnoosh Zamanlooy, Mario Díaz, Shahab Asoodeh
2025 J jnl
Lower Bounds on the MMSE of Adversarially Inferring Sensitive Features.
CoRR
Monica Welfert, Nathan Stromberg, Mario Díaz, Lalitha Sankar
2025 B conf
Tensorization of $f$-Divergences.
ISIT
Rodrigo Cruz, Mario Díaz, Flávio P. Calmon
2024 B conf
$\mathrm{E}_{\gamma}$-Mixing Time.
ISIT
Behnoosh Zamanlooy, Shahab Asoodeh, Mario Díaz, Flávio P. Calmon
2024 J jnl
Locally Private Sampling with Public Data.
CoRR
Behnoosh Zamanlooy, Mario Díaz, Shahab Asoodeh
2024 B conf
On the Privacy Guarantees of Differentially Private Stochastic Gradient Descent.
ISIT
Shahab Asoodeh, Mario Díaz
2023 B conf
On the Inevitability of the Rashomon Effect.
ISIT
Lucas Monteiro Paes, Rodrigo Cruz, Flávio P. Calmon, Mario Díaz
2023 J jnl
Privacy Loss of Noisy Stochastic Gradient Descent Might Converge Even for Non-Convex Losses.
CoRR
Shahab Asoodeh, Mario Díaz
2022 J jnl
A Tunable Loss Function for Robust Classification: Calibration, Landscape, and Generalization.
IEEE Trans. Inf. Theory
Tyler Sypherd, Mario Díaz, John Kevin Cava, Gautam Dasarathy, Peter Kairouz, Lalitha Sankar
2021 J jnl
Lower Bounds for the Minimum Mean-Square Error via Neural Network-based Estimation.
CoRR
Mario Díaz, Peter Kairouz, Lalitha Sankar
2021 B conf
Neural Network-based Estimation of the MMSE.
ISIT
Mario Díaz, Peter Kairouz, Jiachun Liao, Lalitha Sankar
2021 B conf
The Impact of Split Classifiers on Group Fairness.
ISIT
Hao Wang, Hsiang Hsu, Mario Díaz, Flávio P. Calmon
2021 J jnl
To Split or not to Split: The Impact of Disparate Treatment in Classification.
IEEE Trans. Inf. Theory
Hao Wang, Hsiang Hsu, Mario Díaz, Flávio P. Calmon
2020 J jnl
On the Robustness of Information-Theoretic Privacy Measures and Mechanisms.
IEEE Trans. Inf. Theory
Mario Díaz, Hao Wang, Flávio P. Calmon, Lalitha Sankar
2020 J jnl
On the alpha-loss Landscape in the Logistic Model.
CoRR
Tyler Sypherd, Mario Díaz, Lalitha Sankar, Gautam Dasarathy
2020 B conf
On the α-loss Landscape in the Logistic Model.
ISIT
Tyler Sypherd, Mario Díaz, Lalitha Sankar, Gautam Dasarathy
2020 B conf
Privacy Amplification of Iterative Algorithms via Contraction Coefficients.
ISIT
Shahab Asoodeh, Mario Díaz, Flávio P. Calmon
2020 J jnl
Privacy Amplification of Iterative Algorithms via Contraction Coefficients.
CoRR
Shahab Asoodeh, Mario Díaz, Flávio P. Calmon
2020 J jnl
Privacy Analysis of Online Learning Algorithms via Contraction Coefficients.
CoRR
Shahab Asoodeh, Mario Díaz, Flávio P. Calmon
2020 J jnl
To Split or Not to Split: The Impact of Disparate Treatment in Classification.
CoRR
Hao Wang, Hsiang Hsu, Mario Díaz, Flávio P. Calmon
2019 B conf
A Tunable Loss Function for Binary Classification.
ISIT
Tyler Sypherd, Mario Díaz, Lalitha Sankar, Peter Kairouz
2019 J jnl
A Tunable Loss Function for Binary Classification.
CoRR
Tyler Sypherd, Mario Díaz, Lalitha Sankar, Peter Kairouz
2019 J jnl
A Tunable Loss Function for Classification.
CoRR
Tyler Sypherd, Mario Díaz, Harshit Laddha, Lalitha Sankar, Peter Kairouz, Gautam Dasarathy
2019 B conf
An Information-Theoretic View of Generalization via Wasserstein Distance.
ISIT
Hao Wang, Mario Díaz, José Cândido Silveira Santos Filho, Flávio P. Calmon
2019 J jnl
Estimation Efficiency Under Privacy Constraints.
IEEE Trans. Inf. Theory
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2019 J jnl
Theoretical Guarantees for Model Auditing with Finite Adversaries.
CoRR
Mario Díaz, Peter Kairouz, Jiachun Liao, Lalitha Sankar
2018 J jnl
On the Contractivity of Privacy Mechanisms.
CoRR
Mario Díaz, Lalitha Sankar, Peter Kairouz
2018 J jnl
On the Noise-Information Separation of a Private Principal Component Analysis Scheme.
CoRR
Mario Díaz, Shahab Asoodeh, Fady Alajaji, Tamás Linder, Serban Teodor Belinschi, James A. Mingo
2018 J jnl
On the Robustness of Information-Theoretic Privacy Measures and Mechanisms.
CoRR
Mario Díaz, Hao Wang, Flávio P. Calmon, Lalitha Sankar
2018 B conf
The Utility Cost of Robust Privacy Guarantees.
ISIT
Hao Wang, Mario Díaz, Flávio P. Calmon, Lalitha Sankar
2018 J jnl
The Utility Cost of Robust Privacy Guarantees.
CoRR
Hao Wang, Mario Díaz, Flávio P. Calmon, Lalitha Sankar
2017 J jnl
Estimation Efficiency Under Privacy Constraints.
CoRR
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2017 J jnl
On the Capacity of Block Multiantenna Channels.
IEEE Trans. Inf. Theory
Mario Díaz, Víctor Pérez-Abreu
2017 J jnl
Privacy-Aware Guessing Efficiency.
CoRR
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2017 B conf
Privacy-aware guessing efficiency.
ISIT
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2016 J jnl
Information Extraction Under Privacy Constraints.
Inf.
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2016 J jnl
On the Symmetries and the Capacity Achieving Input Covariance Matrices of Multiantenna Channels.
CoRR
Mario Díaz
2016 B conf
On the symmetries and the capacity achieving input covariance matrices of multiantenna channels.
ISIT
Mario Díaz
2015 J jnl
Information Extraction Under Privacy Constraints.
CoRR
Shahab Asoodeh, Mario Díaz, Fady Alajaji, Tamás Linder
2015 J jnl
On the Capacity of Block Multiantenna Channels.
CoRR
Mario Díaz, Víctor Pérez-Abreu
2014 J jnl
On the Asymptotic Ergodic Capacity of Correlated Multiantenna Channels using Finite Dimensional Statistics.
CoRR
Mario Díaz, Víctor Pérez-Abreu