Research map / Federated learning and privacy / Differential privacy
Exact Optimality of Communication-Privacy-Utility Tradeoffs in Distributed Mean Estimation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use RRSC instead of SQKR | Mean estimation under communication and ε-LDP constraints · Simulated distribution-free mean estimation | ℓ2 error | Helped |
| Use RRSC instead of MMRC | Mean estimation under communication and ε-LDP constraints · Simulated distribution-free mean estimation | ℓ2 error | Helped |
| Use RRSC with a small communication budget instead of PrivUnitG without a communication constraint | Mean estimation · Simulated distribution-free mean estimation | ℓ2 error | No clear effect |
| Increase the communication budget for the k-closest scheme | RRSC · Mean estimation under ε-LDP | MSE | Helped |
| Use a rotationally symmetric codebook rather than an arbitrary codebook (ablation) | Shared-randomness mean estimation mechanism · Worst-case mean estimation under ε-LDP | worst-case MSE | Helped |
| Use k-closest probability assignment rather than another ε-LDP assignment (ablation) | Randomly rotating simplex codebook · Worst-case mean estimation under ε-LDP | MSE | Helped |
Closest papers by meaning
- Universal Exact Compression of Differentially Private Mechanisms (NeurIPS 2024 · similarity 0.90)
- Private Vector Mean Estimation in the Shuffle Model: Optimal Rates Require Many Messages (ICML 2024 · similarity 0.89)
- Improved Communication-Privacy Trade-offs in $L_2$ Mean Estimation under Streaming Differential Privacy (ICML 2024 · similarity 0.89)
- A Huber Loss Minimization Approach to Mean Estimation under User-level Differential Privacy (NeurIPS 2024 · similarity 0.87)
- The Privacy Power of Correlated Noise in Decentralized Learning (ICML 2024 · similarity 0.87)
- Exactly Minimax-Optimal Locally Differentially Private Sampling (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.