## Privacy 2

Moderator: Matt J. Kusner

Thu 22 Jul 7 a.m. PDT — 8 a.m. PDT

Abstract:

### Chat is not available.

Thu 22 July 7:00 - 7:20 PDT

(Oral)
##### Locally Private k-Means in One Round

Alisa Chang · Badih Ghazi · Ravi Kumar · Pasin Manurangsi

We provide an approximation algorithm for k-means clustering in the \emph{one-round} (aka \emph{non-interactive}) local model of differential privacy (DP). Our algorithm achieves an approximation ratio arbitrarily close to the best \emph{non private} approximation algorithm, improving upon previously known algorithms that only guarantee large (constant) approximation ratios. Furthermore, ours is the first constant-factor approximation algorithm for k-means that requires only \emph{one} round of communication in the local DP model, positively resolving an open question of Stemmer (SODA 2020). Our algorithmic framework is quite flexible; we demonstrate this by showing that it also yields a similar near-optimal approximation algorithm in the (one-round) shuffle DP model.

[ Paper ]
[ ]
Thu 22 July 7:20 - 7:25 PDT

(Spotlight)
##### Matrix Sketching for Secure Collaborative Machine Learning

Mengjiao Zhang · Shusen Wang

Collaborative learning allows participants to jointly train a model without data sharing. To update the model parameters, the central server broadcasts model parameters to the clients, and the clients send updating directions such as gradients to the server. While data do not leave a client device, the communicated gradients and parameters will leak a client's privacy. Attacks that infer clients' privacy from gradients and parameters have been developed by prior work. Simple defenses such as dropout and differential privacy either fail to defend the attacks or seriously hurt test accuracy. We propose a practical defense which we call Double-Blind Collaborative Learning (DBCL). The high-level idea is to apply random matrix sketching to the parameters (aka weights) and re-generate random sketching after each iteration. DBCL prevents clients from conducting gradient-based privacy inferences which are the most effective attacks. DBCL works because from the attacker's perspective, sketching is effectively random noise that outweighs the signal. Notably, DBCL does not much increase computation and communication costs and does not hurt test accuracy at all.

[ Paper ]
[ ]
Thu 22 July 7:25 - 7:30 PDT

(Spotlight)
##### Markpainting: Adversarial Machine Learning meets Inpainting

David G Khachaturov · Ilia Shumailov · Yiren Zhao · Nicolas Papernot · Ross Anderson

Inpainting is a learned interpolation technique that is based on generative modeling and used to populate masked or missing pieces in an image; it has wide applications in picture editing and retouching. Recently, inpainting started being used for watermark removal, raising concerns. In this paper we study how to manipulate it using our markpainting technique. First, we show how an image owner with access to an inpainting model can augment their image in such a way that any attempt to edit it using that model will add arbitrary visible information. We find that we can target multiple different models simultaneously with our technique. This can be designed to reconstitute a watermark if the editor had been trying to remove it. Second, we show that our markpainting technique is transferable to models that have different architectures or were trained on different datasets, so watermarks created using it are difficult for adversaries to remove. Markpainting is novel and can be used as a manipulation alarm that becomes visible in the event of inpainting. Source code is available at: https://github.com/iliaishacked/markpainting.

[ Paper ]
[ ]
Thu 22 July 7:30 - 7:35 PDT

(Spotlight)
##### Differentially-Private Clustering of Easy Instances

Edith Cohen · Haim Kaplan · Yishay Mansour · Uri Stemmer · Eliad Tsfadia

Clustering is a fundamental problem in data analysis. In differentially private clustering, the goal is to identify k cluster centers without disclosing information on individual data points. Despite significant research progress, the problem had so far resisted practical solutions. In this work we aim at providing simple implementable differentrially private clustering algorithms when the the data is "easy," e.g., when there exists a significant separation between the clusters.

For the easy instances we consider, we have a simple implementation based on utilizing non-private clustering algorithms, and combining them privately. We are able to get improved sample complexity bounds in some cases of Gaussian mixtures and k-means. We complement our theoretical algorithms with experiments of simulated data.

[ Paper ]
[ ]
Thu 22 July 7:35 - 7:40 PDT

(Spotlight)
##### Inference for Network Regression Models with Community Structure

Mengjie Pan · Tyler Mccormick · Bailey Fosdick

Network regression models, where the outcome comprises the valued edge in a network and the predictors are actor or dyad-level covariates, are used extensively in the social and biological sciences. Valid inference relies on accurately modeling the residual dependencies among the relations. Frequently homogeneity assumptions are placed on the errors which are commonly incorrect and ignore critical natural clustering of the actors. In this work, we present a novel regression modeling framework that models the errors as resulting from a community-based dependence structure and exploits the subsequent exchangeability properties of the error distribution to obtain parsimonious standard errors for regression parameters.

[ Paper ]
[ ]
Thu 22 July 7:40 - 7:45 PDT

(Spotlight)
##### DeepReDuce: ReLU Reduction for Fast Private Inference

Nandan Kumar Jha · Zahra Ghodsi · Siddharth Garg · Brandon Reagen

The recent rise of privacy concerns has led researchers to devise methods for private neural inference---where inferences are made directly on encrypted data, never seeing inputs. The primary challenge facing private inference is that computing on encrypted data levies an impractically-high latency penalty, stemming mostly from non-linear operators like ReLU. Enabling practical and private inference requires new optimization methods that minimize network ReLU counts while preserving accuracy. This paper proposes DeepReDuce: a set of optimizations for the judicious removal of ReLUs to reduce private inference latency. The key insight is that not all ReLUs contribute equally to accuracy. We leverage this insight to drop, or remove, ReLUs from classic networks to significantly reduce inference latency and maintain high accuracy. Given a network architecture, DeepReDuce outputs a Pareto frontier of networks that tradeoff the number of ReLUs and accuracy. Compared to the state-of-the-art for private inference DeepReDuce improves accuracy and reduces ReLU count by up to 3.5% (iso-ReLU count) and 3.5x (iso-accuracy), respectively.

[ Paper ]
[ ]
Thu 22 July 7:45 - 7:50 PDT

(Spotlight)
##### Label-Only Membership Inference Attacks

Christopher Choquette-Choo · Florian Tramer · Nicholas Carlini · Nicolas Papernot

Membership inference is one of the simplest privacy threats faced by machine learning models that are trained on private sensitive data. In this attack, an adversary infers whether a particular point was used to train the model, or not, by observing the model's predictions. Whereas current attack methods all require access to the model's predicted confidence score, we introduce a label-only attack that instead evaluates the robustness of the model's predicted (hard) labels under perturbations of the input, to infer membership. Our label-only attack is not only as-effective as attacks requiring access to confidence scores, it also demonstrates that a class of defenses against membership inference, which we call confidence masking'' because they obfuscate the confidence scores to thwart attacks, are insufficient to prevent the leakage of private information. Our experiments show that training with differential privacy or strong L2 regularization are the only current defenses that meaningfully decrease leakage of private information, even for points that are outliers of the training distribution.

[ Paper ]
[ ]
Thu 22 July 7:50 - 7:55 PDT

(Q&A)

[ ]