Utility/privacy trade-off as regularized optimal transport
From MaRDI portal
(Redirected from Publication:6201932)
Abstract: Strategic information is valuable either by remaining private (for instance if it is sensitive) or, on the other hand, by being used publicly to increase some utility. These two objectives are antagonistic and leaking this information might be more rewarding than concealing it. Unlike classical solutions that focus on the first point, we consider instead agents that optimize a natural trade-off between both objectives. We formalize this as an optimization problem where the objective mapping is regularized by the amount of information revealed to the adversary (measured as a divergence between the prior and posterior on the private knowledge). Quite surprisingly, when combined with the entropic regularization, the Sinkhorn loss naturally emerges in the optimization objective, making it efficiently solvable. We apply these techniques to preserve some privacy in online repeated auctions.
Recommendations
Cites work
- A Lagrangian scheme à la Brenier for the incompressible Euler equations
- Algorithms for the Assignment and Transportation Problems
- An introduction to continuous optimization for imaging
- Business cycle dynamics under rational inattention
- Convex analysis approach to d. c. programming: Theory, algorithms and applications
- Coordinate descent algorithms
- Coresets, sparse greedy approximation, and the Frank-Wolfe algorithm
- DC programming: overview.
- Diagonal Equivalence to Matrices with Prescribed Row and Column Sums
- Discretization of functionals involving the Monge-Ampère operator
- Distance makes the types grow stronger: a calculus for differential privacy
- Fast homomorphic evaluation of deep discretized neural networks
- First-order methods of smooth convex optimization with inexact oracle
- Global optimization with polynomials and the problem of moments
- scientific article; zbMATH DE number 3173999 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- Hybrid Block Successive Approximation for One-Sided Non-Convex Min-Max Problems: Algorithms and Applications
- Introduction to global optimization exploiting space-filling curves
- Introduction to global optimization.
- Introduction to nonlinear and global optimization
- Lagrangian discretization of crowd motion and linear diffusion
- Local matching indicators for transport problems with concave costs
- Minimal Geodesics Along Volume-Preserving Maps, Through Semidiscrete Optimal Transport
- On the Foundations of Quantitative Information Flow
- Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
- Repeated games with incomplete information. With the collaboration of Richard E. Stearns
- Signature classes of transportation polytopes
- Some comments on Wolfe's ‘away step’
- Subgradient methods for saddle-point problems
- The Sinkhorn–Knopp Algorithm: Convergence and Applications
- The Variational Formulation of the Fokker--Planck Equation
- Theoretical Insights Into the Optimization Landscape of Over-Parameterized Shallow Neural Networks
- Theory of Cryptography
- Transportation problems which can be solved by the use of hirsch-paths for the dual problems
This page was built for publication: Utility/privacy trade-off as regularized optimal transport
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6201932)
