Workshop
Privacy Preserving Machine Learning - PriML and PPML Joint Edition
Borja Balle · James Bell · Aurélien Bellet · Kamalika Chaudhuri · Adria Gascon · Antti Honkela · Antti Koskela · Casey Meehan · Olga Ohrimenko · Mi Jung Park · Mariana Raykova · Mary Anne Smart · Yu-Xiang Wang · Adrian Weller
Fri 11 Dec, 1:20 a.m. PST
This one day workshop focuses on privacy preserving techniques for machine learning and disclosure in large scale data analysis, both in the distributed and centralized settings, and on scenarios that highlight the importance and need for these techniques (e.g., via privacy attacks). There is growing interest from the Machine Learning (ML) community in leveraging cryptographic techniques such as Multi-Party Computation (MPC) and Homomorphic Encryption (HE) for privacy preserving training and inference, as well as Differential Privacy (DP) for disclosure. Simultaneously, the systems security and cryptography community has proposed various secure frameworks for ML. We encourage both theory and application-oriented submissions exploring a range of approaches listed below. Additionally, given the tension between the adoption of machine learning technologies and ethical, technical and regulatory issues about privacy, as highlighted during the COVID-19 pandemic, we invite submissions for the special track on this topic.
Schedule
Fri 1:20 a.m. - 1:30 a.m.
|
Welcome & Introduction
(
Live Intro
)
>
|
🔗 |
Fri 1:30 a.m. - 2:00 a.m.
|
Invited Talk #1: Reza Shokri (National University of Singapore)
(
Invited Talk
)
>
SlidesLive Video |
Reza Shokri 🔗 |
Fri 2:00 a.m. - 2:30 a.m.
|
Invited Talk #2: Katrina Ligett (Hebrew University)
(
Invited Talk
)
>
SlidesLive Video |
Katrina Ligett 🔗 |
Fri 2:30 a.m. - 3:00 a.m.
|
Invited Talk Q&A with Reza and Katrina
(
Q&A Session
)
>
|
🔗 |
Fri 3:00 a.m. - 3:10 a.m.
|
Break
|
🔗 |
Fri 3:10 a.m. - 3:25 a.m.
|
Contributed Talk #1: POSEIDON: Privacy-Preserving Federated Neural Network Learning
(
Oral
)
>
SlidesLive Video |
Sinem Sav 🔗 |
Fri 3:25 a.m. - 3:30 a.m.
|
Contributed Talk Q&A
(
Q&A Session
)
>
|
🔗 |
Fri 3:30 a.m. - 5:00 a.m.
|
Poster Session & Social on Gather.Town ( Poster Session ) > link | 🔗 |
Fri 8:30 a.m. - 8:40 a.m.
|
Welcome & Introduction
(
Live Intro
)
>
|
🔗 |
Fri 8:40 a.m. - 9:00 a.m.
|
Invited Talk #3: Carmela Troncoso (EPFL)
(
Invited Talk
)
>
SlidesLive Video |
Carmela Troncoso 🔗 |
Fri 9:00 a.m. - 9:30 a.m.
|
Invited Talk #4: Dan Boneh (Stanford University)
(
Invited Talk
)
>
SlidesLive Video |
Dan Boneh 🔗 |
Fri 9:30 a.m. - 10:00 a.m.
|
Invited Talk Q&A with Carmela and Dan
(
Q&A Session
)
>
|
🔗 |
Fri 10:00 a.m. - 10:10 a.m.
|
Break
|
🔗 |
Fri 10:10 a.m. - 11:10 a.m.
|
Poster Session & Social on Gather.Town ( Poster Session ) > link | 🔗 |
Fri 11:10 a.m. - 11:20 a.m.
|
Break
|
🔗 |
Fri 11:20 a.m. - 11:35 a.m.
|
Contributed Talk #2: On the (Im)Possibility of Private Machine Learning through Instance Encoding
(
Oral
)
>
|
Nicholas Carlini 🔗 |
Fri 11:35 a.m. - 11:50 a.m.
|
Contributed Talk #3: Poirot: Private Contact Summary Aggregation
(
Oral
)
>
SlidesLive Video |
Chenghong Wang 🔗 |
Fri 11:50 a.m. - 12:05 p.m.
|
Contributed Talk #4: Greenwoods: A Practical Random Forest Framework for Privacy Preserving Training and Prediction
(
Oral
)
>
SlidesLive Video |
Harsh Chaudhari 🔗 |
Fri 12:05 p.m. - 12:20 p.m.
|
Contributed Talks Q&A
(
Q&A Session
)
>
|
🔗 |
Fri 12:20 p.m. - 12:25 p.m.
|
Break
|
🔗 |
Fri 12:25 p.m. - 12:40 p.m.
|
Contributed Talk #5: Shuffled Model of Federated Learning: Privacy, Accuracy, and Communication Trade-offs
(
Oral
)
>
SlidesLive Video |
Deepesh Data 🔗 |
Fri 12:40 p.m. - 12:55 p.m.
|
Contributed Talk #6: Sample-efficient proper PAC learning with approximate differential privacy
(
Oral
)
>
SlidesLive Video |
Badih Ghazi 🔗 |
Fri 12:55 p.m. - 1:10 p.m.
|
Contributed Talk #7: Training Production Language Models without Memorizing User Data
(
Oral
)
>
SlidesLive Video |
Swaroop Ramaswamy · Om Thakkar 🔗 |
Fri 1:10 p.m. - 1:25 p.m.
|
Contributed Talks Q&A
(
Q&A Session
)
>
|
🔗 |
-
|
Towards General-purpose Infrastructure for Protecting Scientific Data Under Study
(
Poster
)
>
|
Kritika Prakash 🔗 |
-
|
Robust and Private Learning of Halfspaces
(
Poster
)
>
SlidesLive Video |
Badih Ghazi 🔗 |
-
|
Randomness Beyond Noise: Differentially Private Optimization Improvement through Mixup
(
Poster
)
>
SlidesLive Video |
Hanshen Xiao 🔗 |
-
|
Generative Adversarial User Privacy in Lossy Single-Server Information Retrieval
(
Poster
)
>
SlidesLive Video |
Chung-Wei Weng 🔗 |
-
|
Privacy Preserving Chatbot Conversations
(
Poster
)
>
SlidesLive Video |
Debmalya Biswas 🔗 |
-
|
Distributed Differentially Private Averaging with Improved Utility and Robustness to Malicious Parties
(
Poster
)
>
SlidesLive Video |
Aurélien Bellet 🔗 |
-
|
Twinify: A software package for differentially private data release
(
Poster
)
>
SlidesLive Video |
Joonas Jälkö 🔗 |
-
|
DAMS: Meta-estimation of private sketch data structures for differentially private contact tracing
(
Poster
)
>
|
Praneeth Vepakomma 🔗 |
-
|
Secure Medical Image Analysis with CrypTFlow
(
Poster
)
>
SlidesLive Video |
Javier Alvarez-Valle 🔗 |
-
|
Multi-Headed Global Model for handling Non-IID data
(
Poster
)
>
|
Himanshu Arora 🔗 |
-
|
Individual Privacy Accounting via a Rényi Filter
(
Poster
)
>
SlidesLive Video |
Vitaly Feldman 🔗 |
-
|
Does Domain Generalization Provide Inherent Membership Privacy
(
Poster
)
>
SlidesLive Video |
Divyat Mahajan 🔗 |
-
|
Hiding Among the Clones: A Simple and Nearly Optimal Analysis of Privacy Amplification by Shuffling
(
Poster
)
>
SlidesLive Video |
Vitaly Feldman 🔗 |
-
|
SparkFHE: Distributed Dataflow Framework with Fully Homomorphic Encryption
(
Poster
)
>
SlidesLive Video |
Peizhao Hu 🔗 |
-
|
Enabling Fast Differentially Private SGD via Static Graph Compilation and Batch-Level Parallelism
(
Poster
)
>
SlidesLive Video |
Pranav Subramani 🔗 |
-
|
Local Differentially Private Regret Minimization in Reinforcement Learning
(
Poster
)
>
SlidesLive Video |
Evrard Garcelon 🔗 |
-
|
SWIFT: Super-fast and Robust Privacy-Preserving Machine Learning
(
Poster
)
>
SlidesLive Video |
Nishat Koti 🔗 |
-
|
Differentially Private Stochastic Coordinate Descent
(
Poster
)
>
SlidesLive Video |
Georgios Damaskinos 🔗 |
-
|
MP2ML: A Mixed-Protocol Machine LearningFramework for Private Inference
(
Poster
)
>
SlidesLive Video |
Fabian Boemer 🔗 |
-
|
Dataset Inference: Ownership Resolution in Machine Learning
(
Poster
)
>
SlidesLive Video |
Nicolas Papernot 🔗 |
-
|
Privacy-preserving XGBoost Inference
(
Poster
)
>
SlidesLive Video |
Xianrui Meng 🔗 |
-
|
New Challenges for Fully Homomorphic Encryption
(
Poster
)
>
SlidesLive Video |
Marc Joye 🔗 |
-
|
Differentially Private Bayesian Inference For GLMs
(
Poster
)
>
SlidesLive Video |
Joonas Jälkö 🔗 |
-
|
Robustness Threats of Differential Privacy
(
Poster
)
>
|
Ivan Oseledets 🔗 |
-
|
Asymmetric Private Set Intersection with Applications to Contact Tracing and Private Vertical Federated Machine Learning
(
Poster
)
>
SlidesLive Video |
Bogdan Cebere 🔗 |
-
|
Characterizing Private Clipped Gradient Descent on Convex Generalized Linear Problems
(
Poster
)
>
SlidesLive Video |
Shuang Song 🔗 |
-
|
Adversarial Attacks and Countermeasures on Private Training in MPC
(
Poster
)
>
|
Matthew Jagielski 🔗 |
-
|
Optimal Client Sampling for Federated Learning
(
Poster
)
>
SlidesLive Video |
Samuel Horváth 🔗 |
-
|
Data Appraisal Without Data Sharing
(
Poster
)
>
SlidesLive Video |
Mimee Xu 🔗 |
-
|
Dynamic Channel Pruning for Privacy
(
Poster
)
>
|
Abhishek Singh 🔗 |
-
|
Mitigating Leakage in Federated Learning with Trusted Hardware
(
Poster
)
>
SlidesLive Video |
Javad Ghareh Chamani 🔗 |
-
|
Unifying Privacy Loss for Data Analytics
(
Poster
)
>
SlidesLive Video |
Ryan Rogers 🔗 |
-
|
Differentially Private Generative Models Through Optimal Transport
(
Poster
)
>
SlidesLive Video |
Karsten Kreis 🔗 |
-
|
A Principled Approach to Learning Stochastic Representations for Privacy in Deep Neural Inference
(
Poster
)
>
SlidesLive Video |
FatemehSadat Mireshghallah 🔗 |
-
|
Challenges of Differentially Private Prediction in Healthcare Settings
(
Poster
)
>
|
Nicolas Papernot 🔗 |
-
|
Machine Learning with Membership Privacy via Knowledge Transfer
(
Poster
)
>
SlidesLive Video |
Virat Shejwalkar 🔗 |
-
|
Secure Single-Server Aggregation with (Poly)Logarithmic Overhead
(
Poster
)
>
|
James Bell 🔗 |
-
|
PrivAttack: A Membership Inference AttackFramework Against Deep Reinforcement LearningAgents
(
Poster
)
>
SlidesLive Video |
Maziar Gomrokchi 🔗 |
-
|
Effectiveness of MPC-friendly Softmax Replacement
(
Poster
)
>
SlidesLive Video |
Marcel Keller 🔗 |
-
|
Revisiting Membership Inference Under Realistic Assumptions
(
Poster
)
>
|
Bargav Jayaraman 🔗 |
-
|
DYSAN: Dynamically sanitizing motion sensor data against sensitive inferences through adversarial networks
(
Poster
)
>
SlidesLive Video |
Théo JOURDAN 🔗 |
-
|
Fairness in the Eyes of the Data: Certifying Machine-Learning Models
(
Poster
)
>
SlidesLive Video |
Carsten Baum 🔗 |
-
|
Privacy in Multi-armed Bandits: Fundamental Definitions and Lower Bounds on Regret
(
Poster
)
>
SlidesLive Video |
Debabrota Basu 🔗 |
-
|
Accuracy, Interpretability and Differential Privacy via Explainable Boosting
(
Poster
)
>
SlidesLive Video |
Harsha Nori 🔗 |
-
|
Privacy Amplification by Decentralization
(
Poster
)
>
SlidesLive Video |
Aurélien Bellet 🔗 |
-
|
Privacy Risks in Embedded Deep Learning
(
Poster
)
>
SlidesLive Video |
Virat Shejwalkar 🔗 |
-
|
Quantifying Privacy Leakage in Graph Embedding
(
Poster
)
>
SlidesLive Video |
Antoine Boutet 🔗 |
-
|
Understanding Unintended Memorization in Federated Learning
(
Poster
)
>
SlidesLive Video |
Om Thakkar 🔗 |
-
|
Network Generation with Differential Privacy
(
Poster
)
>
SlidesLive Video |
Xu Zheng 🔗 |
-
|
Privacy Regularization: Joint Privacy-Utility Optimization in Language Models
(
Poster
)
>
SlidesLive Video |
FatemehSadat Mireshghallah 🔗 |
-
|
Tight Approximate Differential Privacy for Discrete-Valued Mechanisms Using FFT
(
Poster
)
>
SlidesLive Video |
Antti Koskela 🔗 |
-
|
Differentially private cross-silo federated learning
(
Poster
)
>
SlidesLive Video |
Mikko Heikkilä 🔗 |
-
|
CrypTen: Secure Multi-Party Computation Meets Machine Learning
(
Poster
)
>
|
Shubho Sengupta 🔗 |
-
|
On Polynomial Approximations for Privacy-Preserving and Verifiable ReLU Networks
(
Poster
)
>
|
Salman Avestimehr 🔗 |
-
|
Data-oblivious training for XGBoost models
(
Poster
)
>
SlidesLive Video |
Chester Leung 🔗 |
-
|
Privacy Attacks on Machine Unlearning
(
Poster
)
>
SlidesLive Video |
Ji Gao 🔗 |
-
|
SOTERIA: In Search of Efficient Neural Networks for Private Inference
(
Poster
)
>
SlidesLive Video |
Reza Shokri 🔗 |
-
|
On the Sample Complexity of Privately Learning Unbounded High-Dimensional Gaussians
(
Poster
)
>
SlidesLive Video |
Ishaq Aden-Ali 🔗 |