Overview

• Privacy-preserving data analysis has become essential in the age of Machine Learning (ML) where access to vast amounts of data can provide gains over tuned algorithms. A large proportion of user-contributed data comes from natural language e.g., text transcriptions from voice assistants.

• It is therefore important to curate NLP datasets while preserving the privacy of the users whose data is collected, and train ML models that only retain non-identifying user data.

• The workshop aims to bring together practitioners and researchers from academia and industry to discuss the challenges and approaches to designing, building, verifying, and testing privacy preserving systems in the context of Natural Language Processing.

Agenda

Venue: Virtual

Date: November 20, 2020

Timezone: PST – Pacific Standard Time

08:45 Welcome

• • Oluwaseyi Feyisetan

SESSION 1: 09:00 -- 10:15

09:00 Invited Talk

• • Oracle Efficient Differentially Private Learning and Analysis

  • Aaron Roth (University of Pennsylvania)

09:45 Research Paper

• • On Log-Loss Scores and (No) Privacy [video]

  • Abhinav Aggarwal, Zekun Xu, Oluwaseyi Feyisetan and Nathanael Teissier (Amazon)

SESSION 2: 10:30 -- 12:15

10:30 Invited Talk

• • Data Privacy in Machine Learning

  • Reza Shokri (National University of Singapore)

11:15 Research Paper

• • A Differentially Private Text Perturbation Method Using Regularized Mahalanobis Metric [video]

  • Zekun Xu, Abhinav Aggarwal, Oluwaseyi Feyisetan and Nathanael Teissier (Amazon)

11:45 Research Paper

• • TextHide: Tackling Data Privacy in Language Understanding Tasks [video]

  • Yangsibo Huang, Zhao Song, Danqi Chen, Kai Li and Sanjeev Arora (Princeton University)

12:15 Break

• • Lunch break

SESSION 3: 13:00 -- 14:45

13:00 Invited Talk

• • NLP and Differential Privacy with Metrics

  • Mark Dras and Annabelle McIver (Macquarie University)

13:45 Research Paper

• • Identifying and Classifying Third-party Entities in Natural Language Privacy Policies [video]

  • Mitra Bokaie Hosseini, Pragyan K C, Irwin Reyes and Serge Egelman (St Mary's University)

14:15 Research Paper

• • Surfacing Privacy Settings Using Semantic Matching [video]

  • Rishabh Khandelwal, Asmit Nayak, Yao Yao and Kassem Fawaz (University of Wisconsin–Madison)

SESSION 4: 15:00 -- 17:00

15:00 Invited Talk

• • Privacy in AI/ML Systems: Practical Challenges and Lessons Learned [video]

  • Krishnaram Kenthapadi (Amazon)

15:45 Research Paper

• • Differentially Private Language Models Benefit from Public Pre-training [video]

  • Gavin Kerrigan, Dylan Slack and Jens Tuyls (University of California, Irvine)

16:15 Research Paper

• • A Semantics-based Approach to Disclosure Classification in User-Generated Online Content [video]

  • Chandan Akiti, Anna Squicciarini, Sarah Rajtmajer (Pennsylvania State University)

16:45 Closing remarks

Invited Speakers

Aaron Roth (University of Pennsylvania)

Reza Shokri (National University of Singapore)

Krishnaram Kenthapadi (Amazon AWS)

Annabelle McIver (Macquarie University)

Mark Dras (Macquarie University)

Key Dates

• Submission Deadline: August 28, 2020 September 4, 2020 (11.59pm UTC-12)

• Acceptance Notification: September 25, 2020

• Camera-ready versions: October 10, 2020

• Workshop: November 20, 2020

Contact

privatenlp-emnlp@googlegroups.com