method · first seen 5 Oct 2026
DP-SGD
Also: DPSGD, differentially private SGD, differentially private training, differentially private fine-tuning, differentially private stochastic gradient descent
Stochastic gradient descent with gradient clipping and noise to provide differential privacy guarantees.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Differentially Private Mixing of Public Datasets Improves Private LearningPapers · 5 Oct 2026 · Yufei Chen, Tejumade Afonja, Anvith Thudi, Nicolas Papernot
- Closed-Form Noise Calibration Against Membership Inference for Random-Allocation DP-SGDPapers · 7 Oct 2026 · Murat Bilgehan Ertan, Marten van Dijk
- Better Batches, Better Privacy: Random Allocation for DP-BandMFPapers · 8 Oct 2026 · Ryan McKenna, Arun Ganesh
- From Pixels to Privacy-Sensitive Features: An Information-Theoretic Framework for Reconstruction BoundsPapers · 9 Oct 2026 · Henry Bourne, Robert F. Allison
- Differential Privacy Protected Preference Alignment on Survey Derived DataPapers · 12 Oct 2026 · Xiuyuan Qi, Yi Wang, Wenjie Wang
- Can Rényi Accountants Beat Lossless Accountants for DP-SGD?Papers · 12 Oct 2026 · Jinho Bok
Latest sources
- MOMENT-DP: Momentum Aware Evaluation of Correlated Noise for Differentially Private TrainingPapers · 8 Nov 2026 · Sunny Gupta, Raja Viswanadha
- Exploring the Privacy--Size Tradeoff for Differentially Private Self-Attentive Variational AutoencodersPapers · 3 Nov 2026 · Rohit Sadavarte, Dhruvi Kothari, Sparsh Gupta, Eric Van Oeveren
- Sensitivity Inflation from Cross-Record Precision Routing in Private Quantized Fine-TuningPapers · 1 Nov 2026 · Annan Zhang, Shengjie Song, Wenshuai Yao, Lihao Zhang, Tianmiao
- Counting on Dummies: Privacy Amplification with Observable Counts and Truncated BatchesPapers · 30 Oct 2026 · Vitaly Feldman, Audra McMillan, Kunal Talwar
- Fast Privacy Loss Accounting for Random Allocation via Multiscale FFTPapers · 30 Oct 2026 · Vitaly Feldman, Moshe Shenfeld
- When to spend privacy: optimal allocation in private learningPapers · 29 Oct 2026 · Umberto Biccari, Yongcun Song
- Gaussian Approximation for Composition and Subsampling of Differentially Private AlgorithmsPapers · 23 Oct 2026 · Vitaly Feldman, Moshe Shenfeld
- Curvature-Aware Clipping: Mitigating Bias in DPSGD via Approximate Second-Order InformationPapers · 23 Oct 2026 · Xincheng Xu, Thilina Ranbaduge, Thierry Rakotoarivelo, David Smith
- Unit-Aware Differentially Private Quantum Machine Learning for Clinical and Wearable Health DataPapers · 20 Oct 2026 · Md Khalid Syfullah, Alvi Ataur Khalil
- How Does Muon Work in Private Deep Learning? Scalable Differentially Private Training via OrthogonalizationPapers · 17 Oct 2026 · Alexander Gaponov, Grigory Malinovsky, Egor Shulgin, Peter Richtarik
- Towards Hyperparameter Transfer for Differentially Private OptimizationPapers · 16 Oct 2026 · Tianze Wang, Zhiqi Bu, Linjun Zhang
- Exploring The Empirical Privacy Landscape of Synthetic DataPapers · 15 Oct 2026 · Zihao Zhao, Avi Schwarzschild, Casey Meehan, Katherine Lee, Zhili Feng, Dylan Sa…