Seng Pei Liew

Seng Pei Liew

Research Scientist
Machine Learning
Physics

ICML 2022 privacy papers

1 minute read

Published:

I have curated and am beginning to read ICML ‘22 papers related to privacy and federated learning. The list will be constantly updated with the paper summaries. Stay tuned!

TitleSummary
FedNew: A Communication-Efficient and Privacy-Preserving Newton-Type Method for Federated Learning 
Differentially Private Approximate Quantiles 
A Joint Exponential Mechanism For Differentially Private Top-k 
Transfer Learning In Differential Privacy’s Hybrid-Model 
Bounding Training Data Reconstruction in Private (Deep) Learning 
FriendlyCore: Practical Differentially Private Aggregation 
Public Data-Assisted Mirror Descent for Private Model Training 
Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data 
Selective Network Linearization for Efficient Private Inference 
Shuffle Private Linear Contextual Bandits 
Optimal Algorithms for Mean Estimation under Local Differential Privacy 
Task-aware Privacy Preservation for Multi-dimensional Data 
Differentially Private Coordinate Descent for Composite Empirical Risk Minimization 
Private Streaming SCO in l_p geometry with Applications in High Dimensional Online Decision Making 
Private optimization in the interpolation regime: faster rates and hardness results 
Private Adaptive Optimization with Side information 
The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning 
Faster Privacy Accounting via Evolving DiscretizationImproved numerical accountant
The Poisson Binomial Mechanism for Unbiased Federated Learning with Secure Aggregation 
Private frequency estimation via projective geometry 
Deduplicating Training Data Mitigates Privacy Risks in Language Models 
Hermite Polynomial Features for Private Data Generationuse hermite poly feature instead of random fourier feature for mmd-like generative modeling
Understanding Clipping for Federated Learning: Convergence and Client-Level Differential Privacy 
Differentially Private Community Detection for Stochastic Block Models 
Improved Regret for Differentially Private Exploration in Linear MDP 
Differentially Private Maximal Information Coefficients 
Privacy for Free: How does Dataset Condensation Help Privacy?see this thread
Improved Rates for Differentially Private Stochastic Convex Optimization with Heavy-Tailed Data 
Tight and Robust Private Mean Estimation with Few Users 
Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learning 

You May Also Enjoy

ICLR 2023 privacy/FL papers

3 minute read

Published:

I have curated and am beginning to read ICLR ‘23 papers related to privacy and federated learning. The list will be constantly updated with the paper summaries. Stay tuned!

Awesome list of DP resources

less than 1 minute read

Published:

I am curating a list of resources for learning subjects related to differential privacy, federated learning and machine learning (subject to my preferences).

Privacy papers in ICML 2021

2 minute read

Published:

I have curated and am beginning to read ICML ‘21 papers related to privacy and federated learning. The list will be constantly updated with the paper summaries. Stay tuned!
Note that I wrote a simple script to scrape the links to the paper and the links may not be accurate.

Privacy papers in ICLR 2021

1 minute read

Published:

I have curated and am beginning to read ICLR ‘21 papers related to privacy and federated learning. The list will be constantly updated with the paper summaries. Stay tuned!
Note that I wrote a simple script to scrape the links to the paper and the links may not be accurate.