Research map / Optimization / Nonconvex and smooth optimization
Faster Differentially Private Convex Optimization via Second-Order Methods
Arun Ganesh, Mahdi Haghifam, Thomas Steinke, Abhradeep Guha Thakurta
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 second-order DP solver instead of DP-GD | DP-GD · covertype | run-time ratio T*_DP-GD / T*_ours at eps=0.01: 24.19 \times | Improved |
| use second-order DP solver instead of DP-GD | DP-GD · a1a | run-time ratio at eps=10: 30.59\times | Improved |
| use second-order DP solver instead of DP-GD | DP-GD · synthetic | run-time ratio at eps=0.1: 2.90 \times | Improved |
| increase privacy budget epsilon from 0.01 to 10 | Ours vs DP-GD · adult | run-time ratio: 12.08 \times -> 38.16 \times | Improved |
| use second-order DP solver instead of DP-GD | DP-GD · challenging datasets | run time to same excess loss: $10$-$40\times$ | Improved |
| use damped private Newton method baseline | Ours · several datasets | excess loss | Hurt |
| compare with DP-GD-Oracle (non-private tuned stepsize) | Ours · convergence speed | excess loss | Improved |
| minibatch variant versus DP-SGD | DP-SGD · all datasets | run time to same excess error: $8$-$10 \times $ | Improved |
| use adding operator in minibatch variant | Ours · minibatch | excess loss | Improved |
Closest papers by meaning
- Faster Algorithms for User-Level Private Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.91)
- Improved Rates of Differentially Private Nonconvex-Strongly-Concave Minimax Optimization (AAAI 2025 · similarity 0.90)
- Second-Order Convergence in Private Stochastic Non-Convex Optimization (NeurIPS 2025 · similarity 0.90)
- Private Zeroth-Order Optimization with Public Data (NeurIPS 2025 · similarity 0.90)
- Differential Private Stochastic Optimization with Heavy-tailed Data: Towards Optimal Rates (AAAI 2025 · similarity 0.90)
- Private Stochastic Convex Optimization with Heavy Tails: Near-Optimality from Simple Reductions (NeurIPS 2024 · similarity 0.89)
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.
