Ricardo Silva Carvalho

I am a Computer Science PhD student at Simon Fraser University (SFU) advised by Ke Wang.
The focus of my current research is on Natural Language Processing (NLP) and Differential Privacy (DP).

During the Summer of 2020, I completed an internship on Amazon Web Services as Applied Scientist Intern with the SageMaker Deep Learning Team in Vancouver, Canada, mentored by Theodore Vasiloudis.

From 2012 to 2018, I was a Senior Data Scientist for the Brazilian Government, where for almost a year I worked as Data Science Manager.

I have a M.Sc. in Computer Science from University of Brasilia (UnB), Brazil, where I was advised by Rommel Novaes Carvalho, and a B.Eng. in Computer Engineering from Instituto Tecnologico de Aeronautica (ITA), Brazil, advised by Clovis Torres Fernandes.

Avatar
News
  • Dec 07, 2021: I will be serving as a PC member for the research track of KDD 2022.
  • Dec 01, 2021: Paper "Incorporating Item Frequency for Differentially Private Set Union" was accepted to AAAI 2022. Paper and code will be released soon!
  • Oct 11, 2021: Paper "Differentially Private Ensemble Classifiers for Data Streams" was accepted to WSDM 2022.
  • Aug 19, 2021: I will be serving as a PC member for the Fifteenth International Conference on Web Search and Data Mining (WSDM 2022).
  • Jul 12, 2021: Paper "Training Differentially Private Neural Networks With Lottery Tickets" was accepted to ESORICS 2021.
  • Jul 08, 2021: Paper "TEM: High Utility Metric Differential Privacy on Text" was accepted to the ICML 2021 Workshop on Theory and Practice of Differential Privacy.
  • Jul 01, 2021: Paper "BRR: Preserving Privacy of Text Data Efficiently on Device" was accepted to the ICML 2021 Workshop on Machine Learning for Data.
  • Nov 30, 2020: I will be serving as a PC member for the research track of KDD 2021.
  • Nov 24, 2020: I am giving a talk about differential privacy for the graduate data mining class at Simon Fraser University (SFU).
  • Sep 21, 2020: Our paper "BRR: Preserving Privacy of Text Data Efficiently on Device" received a best paper award at the Shareable NLP Technologies workshop at Amazon's internal ML conference (AMLC 2020)!
  • May 14, 2020: Paper "Differentially Private Top-k Selection via Stability on Unknown Domain" was accepted to UAI 2020!
  • Feb 17, 2020: I will be joining Amazon during this Summer as an Applied Scientist Intern!
  • Dec 11, 2019: I presented my current work on differentially private GANs at the BC AI Student Showcase 2019.
  • Nov 24, 2019: I have been selected as student volunteer for NeurIPS 2019.
  • Oct 01, 2019: Poster on differentially private lottery ticket accepted to Workshop on Machine Learning with Guarantees at NeurIPS 2019!
Recent Research

Please see complete list of publications here.

  • Incorporating Item Frequency for Differentially Private Set Union
    Ricardo Silva Carvalho, Ke Wang, Lovedeep Gondara
    AAAI Conference on Artificial Intelligence (AAAI 2022)

  • Differentially Private Ensemble Classifiers for Data Streams
    Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho
    ACM International WSDM Conference (WSDM 2022)

  • BRR: Preserving Privacy of Text Data Efficiently on Device
    Ricardo Silva Carvalho *, Theodore Vasiloudis, Oluwaseyi Feyisetan --- * Work done during an internship at Amazon
    Workshop on Machine Learning for Data (ML4Data 2021) at ICML 2021
    Workshop on Shareable NLP Technologies at Amazon's Machine Learning Conference (AMLC 2020) - best paper award
    Paper Amazon Science

  • TEM: High Utility Metric Differential Privacy on Text
    Ricardo Silva Carvalho *, Theodore Vasiloudis, Oluwaseyi Feyisetan --- * Work done during an internship at Amazon
    Workshop on Theory and Practice of Differential Privacy (TPDP 2021) at ICML 2021
    Paper Amazon Science

  • Differentially Private Top-k Selection via Stability on Unknown Domain
    Ricardo Silva Carvalho, Ke Wang, Lovedeep Gondara, Chunyan Miao
    Conference on Uncertainty in Artificial Intelligence (UAI 2020)
    Paper Supplementary File Video Code

  • Training Differentially Private Neural Networks With Lottery Tickets
    Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho
    European Symposium on Research in Computer Security (ESORICS 2021)
    Workshop on Machine Learning with Guarantees at NeurIPS 2019
    Extended Abstract

  • Privacy-Preserving Publication of Sensitive Data using Differentially Private GANs
    Ricardo Silva Carvalho
    BC AI Student Showcase 2019
    Poster Report Code

  • A Text Classification Model to Identify Performance Bonds Requirement in Public Bidding Notices.
    Urias Cruz da Cunha, Ricardo Silva Carvalho, Alexandre Zaghetto
    Future of Information and Communication Conference (FICC 2020)
    Paper

Relevant Activities
  • External reviewer or PC member for AI/ML/DM conferences: KDD 2022, AISTATS 2022, WSDM 2022, ICDM 2021, KDD 2021, WSDM 2021, IEEE BigData 2020, KDD 2020, ICDM 2020, ICDE 2020.
  • Only Teaching Assistant (TA) for the Data Mining course at Simon Fraser University (SFU) with more than 80 undergraduate students enrolled.