Workshop
Privacy in Machine Learning (PriML) 2021
Yu-Xiang Wang · Borja Balle · Giovanni Cherubin · Kamalika Chaudhuri · Antti Honkela · Jonathan Lebensold · Casey Meehan · Mi Jung Park · Adrian Weller · Yuqing Zhu
Tue 14 Dec, 1:20 a.m. PST
The goal of our workshop is to bring together privacy experts working in academia and industry to discuss the present and future of technologies that enable machine learning with privacy. The workshop will focus on the technical aspects of privacy research and deployment with invited and contributed talks by distinguished researchers in the area. By design, the workshop should serve as a meeting point for regular NeurIPS attendees interested/working on privacy to meet other parts of the privacy community (security researchers, legal scholars, industry practitioners). The focus this year will include emerging problems such as machine unlearning, privacy-fairness tradeoffs and legal challenges in recent deployments of differential privacy (e.g. that of the US Census Bureau). We will conclude the workshop with a panel discussion titled: “Machine Learning and Privacy in Practice: Challenges, Pitfalls and Opportunities”. A diverse set of panelists will address the challenges faced applying these technologies to the real world. The programme of the workshop will emphasize the diversity of points of view on the problem of privacy. We will also ensure that there is ample time for discussions that encourage networking between researchers, which should result in mutually beneficial new long-term collaborations.
Schedule
Tue 1:20 a.m. - 1:30 a.m.
|
Introduction
(
Opening
)
>
SlidesLive Video |
🔗 |
Tue 1:30 a.m. - 2:00 a.m.
|
Invited talk: Emiliano de Cristofaro (University College London) --- Privacy in Machine Learning -- It's Complicated
(
Invited talk
)
>
SlidesLive Video |
Emiliano De Cristofaro 🔗 |
Tue 2:00 a.m. - 2:15 a.m.
|
Emiliano Q&A
(
Q&A
)
>
|
🔗 |
Tue 2:15 a.m. - 2:30 a.m.
|
Coffee break
(
coffee break
)
>
|
🔗 |
Tue 2:30 a.m. - 2:45 a.m.
|
Differential Privacy via Group Shuffling
(
Contributed talk
)
>
link
SlidesLive Video |
Amir Mohammad Abouei · Clement Canonne 🔗 |
Tue 2:45 a.m. - 3:00 a.m.
|
SoK: Privacy-preserving Clustering (Extended Abstract)
(
Contributed talk
)
>
link
SlidesLive Video |
Helen Möllering · Hossein Yalame · Thomas Schneider · Aditya Hegde 🔗 |
Tue 3:00 a.m. - 3:15 a.m.
|
Contributed talk Q&A
(
Q&A
)
>
SlidesLive Video |
🔗 |
Tue 3:15 a.m. - 3:30 a.m.
|
Coffee Break
|
🔗 |
Tue 3:30 a.m. - 4:30 a.m.
|
Poster Session ( Poster ) > link | 🔗 |
Tue 4:30 a.m. - 5:15 a.m.
|
Panel
(
Panel
)
>
SlidesLive Video |
Catuscia Palamidessi · Carmela Troncoso · Yang Zhang 🔗 |
Tue 8:20 a.m. - 8:30 a.m.
|
Introduction
(
Opening
)
>
|
🔗 |
Tue 8:30 a.m. - 9:00 a.m.
|
Invited talk: Helen Nissenbaum (Cornell Tech) --- Practical Privacy, Fairness, Ethics, Policy
(
Invited talk
)
>
SlidesLive Video |
Helen Nissenbaum 🔗 |
Tue 9:00 a.m. - 9:30 a.m.
|
Invited talk: Aaron Roth (UPenn / Amazon): Machine Unlearning.
(
Invited talk
)
>
SlidesLive Video |
Aaron Roth 🔗 |
Tue 9:30 a.m. - 10:00 a.m.
|
Q&A for Helen and Aaron
(
Q&A
)
>
SlidesLive Video |
🔗 |
Tue 10:00 a.m. - 10:15 a.m.
|
Coffee break
|
🔗 |
Tue 10:15 a.m. - 11:15 a.m.
|
Poster Session ( Gather.Town ) > link | 🔗 |
Tue 11:15 a.m. - 11:30 a.m.
|
Coffee break
|
🔗 |
Tue 11:30 a.m. - 12:00 p.m.
|
Invited talk: Kristin Lauter (Facebook AI Research): ML on Encrypted Data.
(
Invited talk
)
>
SlidesLive Video |
Kristin E. Lauter 🔗 |
Tue 12:00 p.m. - 12:15 p.m.
|
Q&A for Kristin
(
Q&A
)
>
|
🔗 |
Tue 12:15 p.m. - 12:30 p.m.
|
Privacy-Aware Rejection Sampling
(
Contributed talk
)
>
link
SlidesLive Video |
Jordan Awan · Vinayak Rao 🔗 |
Tue 12:30 p.m. - 12:45 p.m.
|
Population Level Privacy Leakage in Binary Classification wtih Label Noise
(
Contributed talk
)
>
link
SlidesLive Video |
Róbert Busa-Fekete · Andres Munoz Medina · Umar Syed · Sergei Vassilvitskii 🔗 |
Tue 12:45 p.m. - 1:00 p.m.
|
Simple Baselines Are Strong Performers for Differentially Private Natural Language Processing
(
Contributed talk
)
>
link
SlidesLive Video |
Xuechen (Chen) Li · Florian Tramer · Percy Liang · Tatsunori Hashimoto 🔗 |
Tue 1:00 p.m. - 1:15 p.m.
|
Canonical Noise Distributions and Private Hypothesis Tests
(
Contributed talk
)
>
link
SlidesLive Video |
Jordan Awan · Salil Vadhan 🔗 |
Tue 1:15 p.m. - 1:45 p.m.
|
Q&A for four contributed talks
(
Q&A
)
>
SlidesLive Video |
🔗 |
Tue 1:45 p.m. - 2:30 p.m.
|
Panel
(
Panel
)
>
SlidesLive Video |
Oluwaseyi Feyisetan · Helen Nissenbaum · Aaron Roth · Christine Task 🔗 |
Tue 2:30 p.m. - 2:40 p.m.
|
Closing
(
closing
)
>
SlidesLive Video |
🔗 |
-
|
An automatic differentiation system for the age of differential privacy
(
Poster
)
>
SlidesLive Video |
Dmitrii Usynin · Alexander Ziller · Moritz Knolle · Daniel Rueckert · Georgios Kaissis 🔗 |
-
|
Communication Efficient Federated Learning with Secure Aggregation and Differential Privacy
(
Poster
)
>
link
SlidesLive Video |
Wei-Ning Chen · Christopher Choquette-Choo · Peter Kairouz 🔗 |
-
|
Realistic Face Reconstruction from Deep Embeddings
(
Poster
)
>
link
SlidesLive Video |
Edward Vendrow · Joshua Vendrow 🔗 |
-
|
Certified Predictions using MPC-Friendly Publicly Verifiable Covertly Secure Commitments
(
Poster
)
>
link
SlidesLive Video |
Nitin Agrawal · James Bell · Matt Kusner 🔗 |
-
|
Mean Estimation with User-level Privacy under Data Heterogeneity
(
Poster
)
>
link
SlidesLive Video |
Rachel Cummings · Vitaly Feldman · Audra McMillan · Kunal Talwar 🔗 |
-
|
DP-SEP: Differentially private stochastic expectation propagation ( Poster ) > link | Margarita Vinaroz · Mijung Park 🔗 |
-
|
Label Private Deep Learning Training based on Secure Multiparty Computation and Differential Privacy
(
Poster
)
>
link
SlidesLive Video |
Sen Yuan · Milan Shen · Ilya Mironov · Anderson Nascimento 🔗 |
-
|
Private Confidence Sets
(
Poster
)
>
link
SlidesLive Video |
Karan Chadha · John Duchi · Rohith Kuditipudi 🔗 |
-
|
A Joint Exponential Mechanism for Differentially Private Top-k Set
(
Poster
)
>
link
SlidesLive Video |
Andres Munoz Medina · Matthew Joseph · Jennifer Gillenwater · Monica Ribero Diaz 🔗 |
-
|
Adversarial Detection Avoidance Attacks: Evaluating the robustness of perceptual hashing-based client-side scanning
(
Poster
)
>
link
SlidesLive Video |
Shubham Jain · Ana-Maria Cretu · Yves-Alexandre Montjoye 🔗 |
-
|
Characterizing and Improving MPC-based Private Inference for Transformer-based Models
(
Poster
)
>
link
SlidesLive Video |
Yongqin Wang · Brian Knott · Murali Annavaram · Hsien-Hsin Lee 🔗 |
-
|
SoK: Privacy-preserving Clustering (Extended Abstract)
(
Poster
)
>
link
SlidesLive Video |
Helen Möllering · Hossein Yalame · Aditya Hegde · Thomas Schneider 🔗 |
-
|
Membership Inference Attacks Against NLP Classification Models
(
Poster
)
>
link
SlidesLive Video |
Virat Shejwalkar · Huseyin A Inan · Amir Houmansadr · Robert Sim 🔗 |
-
|
A Generic Hybrid 2PC Framework with Application to Private Inference of Unmodified Neural Networks (Extended Abstract)
(
Poster
)
>
link
SlidesLive Video |
Lennart Braun · Thomas Schneider · Rosario Cammarota 🔗 |
-
|
ABY2.0: New Efficient Primitives for STPC with Applications to Privacy in Machine Learning (Extended Abstract)
(
Poster
)
>
link
SlidesLive Video |
Arpita Patra · Hossein Yalame · Thomas Schneider · Ajith Suresh 🔗 |
-
|
Combining Public and Private Data
(
Poster
)
>
link
SlidesLive Video |
Cecilia Ferrando · Jennifer Gillenwater · Alex Kulesza 🔗 |
-
|
Iterative Methods for Private Synthetic Data: Unifying Framework and New Methods
(
Poster
)
>
link
SlidesLive Video |
Terrance Liu · Giuseppe Vietri · Steven Wu 🔗 |
-
|
Unsupervised Membership Inference Attacks Against Machine Learning Models
(
Poster
)
>
link
SlidesLive Video |
YUEFENG PENG 🔗 |
-
|
Population Level Privacy Leakage in Binary Classification wtih Label Noise
(
Poster
)
>
link
SlidesLive Video |
Róbert Busa-Fekete · Andres Munoz Medina · Umar Syed · Sergei Vassilvitskii 🔗 |
-
|
A Novel Self-Distillation Architecture to Defeat Membership Inference Attacks
(
Poster
)
>
link
SlidesLive Video |
Xinyu Tang · Saeed Mahloujifar · Liwei Song · Virat Shejwalkar · Amir Houmansadr · Prateek Mittal 🔗 |
-
|
Enforcing fairness in private federated learning via the modified method of differential multipliers
(
Poster
)
>
link
SlidesLive Video |
Borja Rodríguez Gálvez · Filip Granqvist · Rogier van Dalen · Matthew Seigel 🔗 |
-
|
Efficient passive membership inference attack in federated learning
(
Poster
)
>
link
SlidesLive Video |
CHUAN XU · Giovanni Neglia · Oualid ZARI 🔗 |
-
|
Interaction data are identifiable even across long periods of time
(
Poster
)
>
link
SlidesLive Video |
Ana-Maria Cretu · Federico Monti · Stefano Marrone · Xiaowen Dong · Michael Bronstein · Yves-Alexandre Montjoye 🔗 |
-
|
Simple Baselines Are Strong Performers for Differentially Private Natural Language Processing
(
Poster
)
>
link
SlidesLive Video |
Xuechen (Chen) Li · Florian Tramer · Percy Liang · Tatsunori Hashimoto 🔗 |
-
|
Feature-level privacy loss modelling in differentially private machine learning ( Poster ) > link | Dmitrii Usynin · Alexander Ziller · Moritz Knolle · Daniel Rueckert · Georgios Kaissis 🔗 |
-
|
Opacus: User-Friendly Differential Privacy Library in PyTorch
(
Poster
)
>
link
SlidesLive Video |
11 presentersAshkan Yousefpour · Igor Shilov · Alexandre Sablayrolles · Karthik Prasad · Mani Malek Esmaeili · John Nguyen · Sayan Ghosh · Akash Bharadwaj · Jessica Zhao · Graham Cormode · Ilya Mironov |
-
|
Differential Privacy via Group Shuffling
(
Poster
)
>
link
SlidesLive Video |
Amir Mohammad Abouei · Clement Canonne 🔗 |
-
|
Architecture Matters: Investigating the Influence of Differential Privacy on Neural Network Design
(
Poster
)
>
link
SlidesLive Video |
Felix Morsbach 🔗 |
-
|
Reconstructing Training Data with Informed Adversaries
(
Poster
)
>
link
SlidesLive Video |
Borja Balle · Giovanni Cherubin · Jamie Hayes 🔗 |
-
|
SSSE: Efficiently Erasing Samples from Trained Machine Learning Models ( Poster ) > link | Alexandra Peste · Dan Alistarh · Christoph Lampert 🔗 |
-
|
Differentially Private Hamiltonian Monte Carlo ( Poster ) > link | Ossi Räisä · Antti Koskela · Antti Honkela 🔗 |
-
|
Zero Knowledge Arguments for Verifiable Sampling
(
Poster
)
>
link
SlidesLive Video |
César Sabater · Jan Ramon 🔗 |
-
|
Basil: A Fast and Byzantine-Resilient Approach for Decentralized Training
(
Poster
)
>
link
SlidesLive Video |
Ahmed Elkordy · Saurav Prakash · Salman Avestimehr 🔗 |
-
|
Canonical Noise Distributions and Private Hypothesis Tests
(
Poster
)
>
link
SlidesLive Video |
Jordan Awan · Salil Vadhan 🔗 |
-
|
Privacy-Aware Rejection Sampling
(
Poster
)
>
link
SlidesLive Video |
Jordan Awan · Vinayak Rao 🔗 |
-
|
Reconstructing Test Labels from Noisy Loss Scores (Extended Abstract)
(
Poster
)
>
link
SlidesLive Video |
Abhinav Aggarwal · Shiva Kasiviswanathan · Zekun Xu · Oluwaseyi Feyisetan · Nathanael Teissier 🔗 |
-
|
Understanding Training-Data Leakage from Gradients in Neural Networks for ImageClassifications
(
Poster
)
>
link
SlidesLive Video |
Cangxiong Chen · Neill Campbell 🔗 |
-
|
Sample-and-threshold differential privacy: Histograms and applications
(
Poster
)
>
link
SlidesLive Video |
Graham Cormode 🔗 |
-
|
Tight Accounting in the Shuffle Model of Differential Privacy
(
Poster
)
>
link
SlidesLive Video |
Antti Koskela · Mikko Heikkilä · Antti Honkela 🔗 |