Research map / Optimization / Nonconvex and smooth optimization
Private Stochastic Convex Optimization with Optimal Rates
Raef Bassily, Vitaly Feldman, Kunal Talwar, 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 mini-batch noisy SGD with uniform-stability analysis | DP-SCO with smooth convex losses · Stochastic convex optimization | excess population loss: 10\,ML\cdot\max\left(\frac{\sqrt{d\,\log(1/\delta)}}{\epsilon\, n},~\frac{1}{\sqrt{n}}\right); prior rate: 1/n^{1/4} whe | Improved |
| Apply Moreau-Yosida smoothing before NSGD | DP-SCO with non-smooth convex losses · Stochastic convex optimization (non-smooth losses) | excess population loss: 24\,M\,L\cdot\max\left(\frac{\sqrt{d\,\log(1/\delta)}}{\epsilon\, n},~\frac{1}{\sqrt{n}}\right) | Improved |
| Use approximate proximal operators to compute smoothed-loss gradients | ProxGD (NSGD + Prox) · Stochastic convex optimization (non-smooth losses) | excess population loss: additional term of 2LM/n | No clear effect |
| Add objective perturbation with regularization | DP-SCO under the rank-one Hessian assumption · Stochastic convex optimization (smooth losses) | excess population loss: 2\,M\,L\,\sqrt{\frac{2}{n}+\frac{4\,d\,\log(1/\delta)}{\epsilon^2\, n^2}} | Improved |
| Use approximate objective perturbation with Gaussian noise on the approximate minimizer | DP-SCO under the rank-one Hessian assumption · Stochastic convex optimization (smooth losses) | excess population loss: O\left(M\,L\cdot \max\left(\frac{1}{\sqrt{n}},~ \frac{\sqrt{d\,\log(1/\delta)}}{\epsilon\, n}\right)\right) | No clear effect |
Closest papers by meaning
- Private Stochastic Convex Optimization with Heavy Tails: Near-Optimality from Simple Reductions (NeurIPS 2024 · similarity 0.93)
- Faster Algorithms for User-Level Private Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.93)
- Differential Private Stochastic Optimization with Heavy-tailed Data: Towards Optimal Rates (AAAI 2025 · similarity 0.92)
- Improved Differentially Private and Lazy Online Convex Optimization: Lower Regret without Smoothness Requirements (ICML 2024 · similarity 0.91)
- Public-data Assisted Private Stochastic Optimization: Power and Limitations (NeurIPS 2024 · similarity 0.90)
- Improved Sample Complexity for Private Nonsmooth Nonconvex Optimization (ICML 2025 · similarity 0.90)
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.
