π
π
The Cartographer
Not All Learnable Distribution Classes are Privately Learnable
February 01, 2024 Β· Declared Dead Β· π International Conference on Algorithmic Learning Theory
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Mark Bun, Gautam Kamath, Argyris Mouzakis, Vikrant Singhal
arXiv ID
2402.00267
Category
cs.DS: Data Structures & Algorithms
Cross-listed
cs.CR,
stat.ML
Citations
3
Venue
International Conference on Algorithmic Learning Theory
Last Checked
4 months ago
Abstract
We give an example of a class of distributions that is learnable up to constant error in total variation distance with a finite number of samples, but not learnable under $(\varepsilon, Ξ΄)$-differential privacy with the same target error. This weakly refutes a conjecture of Ashtiani.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Data Structures & Algorithms
R.I.P.
π»
Ghosted
Route Planning in Transportation Networks
R.I.P.
π»
Ghosted
Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration
R.I.P.
π»
Ghosted
Hierarchical Clustering: Objective Functions and Algorithms
R.I.P.
π»
Ghosted
Graph Isomorphism in Quasipolynomial Time
π
π
The Cartographer
Simulation optimization: A review of algorithms and applications
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
π»
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
π»
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
π»
Ghosted