Mirror Descent Based Database Privacy
نویسندگان
چکیده
In this paper, we focus on the problem of private database release in the interactive setting: a trusted database curator receives queries in an online manner for which it needs to respond with accurate but privacy preserving answers. To this end, we generalize the IDC (Iterative Database Construction) framework of [15,13] that maintains a differentially private artificial dataset and answers incoming linear queries using the artificial dataset. In particular, we formulate a generic IDC framework based on the Mirror Descent algorithm, a popular convex optimization algorithm [1]. We then present two concrete applications, namely, cut queries over a bipartite graph and linear queries over lowrank matrices, and provide significantly tighter error bounds than the ones by [15,13].
منابع مشابه
Robust Decentralized Differentially Private Stochastic Gradient Descent
Stochastic gradient descent (SGD) is one of the most applied machine learning algorithms in unreliable large-scale decentralized environments. In this type of environment data privacy is a fundamental concern. The most popular way to investigate this topic is based on the framework of differential privacy. However, many important implementation details and the performance of differentially priv...
متن کاملMirror Descent Search and Acceleration
In recent years, attention has been focused on the relationship between black box optimization and reinforcement learning. Black box optimization is a framework for the problem of finding the input that optimizes the output represented by an unknown function. Reinforcement learning, by contrast, is a framework for finding a policy to optimize the expected cumulative reward from trial and error....
متن کاملFollow-the-Regularized-Leader and Mirror Descent: Equivalence Theorems and L1 Regularization
We prove that many mirror descent algorithms for online convex optimization (such as online gradient descent) have an equivalent interpretation as follow-the-regularizedleader (FTRL) algorithms. This observation makes the relationships between many commonly used algorithms explicit, and provides theoretical insight on previous experimental observations. In particular, even though the FOBOS comp...
متن کاملSparse Q-learning with Mirror Descent
This paper explores a new framework for reinforcement learning based on online convex optimization, in particular mirror descent and related algorithms. Mirror descent can be viewed as an enhanced gradient method, particularly suited to minimization of convex functions in highdimensional spaces. Unlike traditional gradient methods, mirror descent undertakes gradient updates of weights in both t...
متن کاملPotential-Function Proofs for First-Order Methods
This note discusses proofs for convergence of first-order methods based on simple potentialfunction arguments. We cover methods like gradient descent (for both smooth and non-smooth settings), mirror descent, and some accelerated variants.
متن کامل