About Me

Welcome! I’m a Research Scientist in the PRIVATICS team at Inria Grenoble - Rhône-Alpes. Previously, I was a Postdoctoral Researcher in the Comète team at Inria Saclay – LIX, École Polytechnique.

I received my Ph.D. in Computer Science from the University Bourgogne Franche-Comté (UBFC) in 2022, working with Profs. Jean-François Couchot, Bechara Al Bouna, and Xiaokui Xiao.
Before that, I received my M.Eng. in Electrical Engineering from São Paulo State University (UNESP) in 2019, under the supervision of Prof. Aparecido Augusto de Carvalho.

Currently, my research interests are differential privacy and ethical aspects of machine learning (e.g., privacy and fairness issues).

News

• [Oct-23] I started my new position as a research scientist in the PRIVATICS team at Inria Grenoble - Rhône-Alpes.
• [Sep-23] Our paper “Revealing the True Cost of Local Privacy: An Auditing Perspective” has been accepted to TPDP 2023 and can be accessed on https://arxiv.org/abs/2309.01597.
• [Jul-23] Our project “DIFPRIPOS: Making PostgreSQL Differentially Private for Transparent AI” has been accepted for funding by ANR AAPG 2023.
• [Jul-23] Our paper “(Local) Differential Privacy has NO Disparate Impact on Fairness” has received the *Best Paper Award* at DBSec 2023!
• [May-23] We have two papers accepted at DBSec 2023:
• • On the Utility Gain of Iterative Bayesian Update for Locally Differentially Private Mechanisms.
• • (Local) Differential Privacy has NO Disparate Impact on Fairness.
• [Apr-23] I’ll be serving as a Reviewer for the 37th Conference on Neural Information Processing Systems (NeurIPS 2023).
• [Mar-23] I’ll be serving on the PC and Editorial Board of the 24th Privacy Enhancing Technologies Symposium (PETS 2024).
• [Mar-23] I’ll be serving in the PC (Research Track) of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD 2023).
• [Feb-23] I’m serving as a Reviewer and Area Chair (Tiny Paper Track) for the 11th International Conference on Learning Representations (ICLR 2023).
• [Jan-23] I’m serving in the PC of the 6th ACM Conference on Fairness, Accountability, and Transparency (FAccT 2023).
• [Dec-22] Our paper “On the Risks of Collecting Multidimensional Data Under Local Differential Privacy” has been accepted to VLDB 2023 and can be accessed on https://arxiv.org/abs/2209.01684.
• [Dec-22] Our paper “Frequency Estimation of Evolving Data Under Local Differential Privacy” has been accepted to EDBT 2023 and can be accessed on https://arxiv.org/abs/2210.00262.
• [Nov-22] I’m serving in the PC of the 4th AAAI Workshop on Privacy-Preserving Artificial Intelligence (PPAI 2023).
• [Nov-22] I’m serving in the PC of the 10th IEEE Swiss Conference on Data Science (SDS 2023).
• [Oct-22] I gave a seminar talk at the GT-PVP entitled “Locally differentially private protocols for frequency estimation of longitudinal data”.
• [Oct-22] I’ve visited the SYSTOPIA lab at The University of British Columbia in Vancouver for two months (mid October to mid December 2022).
• [Sep-22] Our demonstration paper “Multi-Freq-LDPy: Multiple Frequency Estimation Under Local Differential Privacy in Python” has been accepted to ESORICS 2022 (poster presentation) and can be accessed at https://arxiv.org/abs/2205.02648.
• [Sep-22] We are organizing a Workshop on issues in Ethical Machine Learning with special focus on Privacy and Fairness on the 30th of September. The topics will range from Fairness to Privacy, you can register here.
• [Jul-22] Our paper “Improving the utility of locally differentially private protocols for longitudinal and multidimensional frequency estimates” has been accepted to Digital Communications and Networks journal and can be accessed on https://doi.org/10.1016/j.dcan.2022.07.003.
• [May-22] We’ve realeased a Python package multi-freq-ldpy for multiple frequency estimation under local differential privacy. Please check out the introductory Video_Presentation, Slide_Presentation, and arXived Demonstration Paper of our package.
• [May-22] Our paper “Differentially Private Multivariate Time Series Forecasting of Aggregated Human Mobility With Deep Learning: Input or Gradient Perturbation?” has been accepted to Neural Computing and Applications and can be accessed on https://arxiv.org/abs/2205.00436.
• [Apr-22] I’m a teaching assistant for Introduction to Computer Science with Java (INF361) at École Polytechnique this semester.
• [Apr-22] The final version of my thesis “Production of Categorical Data Verifying Differential Privacy: Conception and Applications to Machine Learning” can be accessed at https://arxiv.org/abs/2204.00850.
• [Apr-22] I’m serving in the PC of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2022).
• [Jan-22] I successfully defended my Ph.D. “Production of Categorical Data Verifying Differential Privacy: Conception and Applications to Machine Learning”. Here are the Presentation and the Video-Presentation. I thank my dissertation jury Profs. Benjamin Nguyen, Mathieu Cunche, Stéphane Chrétien, and Mário S. Alvim.

Contact

I am always happy to discuss the possibility of new collaborations.

• Email: heber.hwang-arcolezi@inria.fr
• Postal Address: 655 Av. de l’Europe, 38330 Montbonnot-Saint-Martin, France.

Last update: 2nd October 2023.