Related papers: More Efficient Privacy Amplification with Less Ran…
We study how inherent randomness in the training process -- where each sample (or client in federated learning) contributes only to a randomly selected portion of training -- can be leveraged for privacy amplification. This includes (1)…
An attractive approach for fast search in image databases is binary hashing, where each high-dimensional, real-valued image is mapped onto a low-dimensional, binary vector and the search is done in this binary space. Finding the optimal…
Extended differential privacy, a generalization of standard differential privacy (DP) using a general metric, has been widely studied to provide rigorous privacy guarantees while keeping high utility. However, existing works on extended DP…
We show that the tasks of privacy amplification against quantum adversaries and data compression with quantum side information are dual in the sense that the ability to perform one implies the ability to perform the other. These are two of…
Deep neural networks are increasingly being used in a variety of machine learning applications applied to rich user data on the cloud. However, this approach introduces a number of privacy and efficiency challenges, as the cloud operator…
Recent work of Erlingsson, Feldman, Mironov, Raghunathan, Talwar, and Thakurta [EFMRTT19] demonstrates that random shuffling amplifies differential privacy guarantees of locally randomized data. Such amplification implies substantially…
Data mining has made broad significant multidisciplinary field used in vast application domains and extracts knowledge by identifying structural relationship among the objects in large data bases. Privacy preserving data mining is a new…
Randomized response is attractive for privacy preserving data collection because the provided privacy can be quantified by means such as differential privacy. However, recovering and analyzing statistics involving multiple dependent…
In this paper, we develop compositional methods for formally verifying differential privacy for algorithms whose analysis goes beyond the composition theorem. Our methods are based on the observation that differential privacy has deep…
A basic problem in the design of privacy-preserving algorithms is the private maximization problem: the goal is to pick an item from a universe that (approximately) maximizes a data-dependent function, all under the constraint of…
The *shuffle model* is a powerful tool to amplify the privacy guarantees of the *local model* of differential privacy. In contrast to the fully decentralized manner of guaranteeing privacy in the local model, the shuffle model requires a…
This paper considers the problem of outsourcing the multiplication of two private and sparse matrices to untrusted workers. Secret sharing schemes can be used to tolerate stragglers and guarantee information-theoretic privacy of the…
Developing a differentially private deep learning algorithm is challenging, due to the difficulty in analyzing the sensitivity of objective functions that are typically used to train deep neural networks. Many existing methods resort to the…
We study the fundamental problem of the construction of optimal randomization in Differential Privacy. Depending on the clipping strategy or additional properties of the processing function, the corresponding sensitivity set theoretically…
Supervised hashing methods are widely-used for nearest neighbor search in computer vision applications. Most state-of-the-art supervised hashing approaches employ batch-learners. Unfortunately, batch-learning strategies can be inefficient…
Large Transformer models routinely achieve state-of-the-art results on a number of tasks but training these models can be prohibitively costly, especially on long sequences. We introduce two techniques to improve the efficiency of…
Hyperplane hashing aims at rapidly searching nearest points to a hyperplane, and has shown practical impact in scaling up active learning with SVMs. Unfortunately, the existing randomized methods need long hash codes to achieve reasonable…
We consider privacy amplification against quantum side information by using regular random binning as an effective extractor. For constant-type sources, we obtain error exponent and strong converse bounds in terms of the so-called quantum…
This work examines a novel question: how much randomness is needed to achieve local differential privacy (LDP)? A motivating scenario is providing {\em multiple levels of privacy} to multiple analysts, either for distribution or for…
Due to the impressive learning power, deep learning has achieved a remarkable performance in supervised hash function learning. In this paper, we propose a novel asymmetric supervised deep hashing method to preserve the semantic structure…