REVIEW 27 cited by
Opacus: User-Friendly Differential Privacy Library in PyTorch
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We introduce Opacus, a free, open-source PyTorch library for training deep learning models with differential privacy (hosted at opacus.ai). Opacus is designed for simplicity, flexibility, and speed. It provides a simple and user-friendly API, and enables machine learning practitioners to make a training pipeline private by adding as little as two lines to their code. It supports a wide variety of layers, including multi-head attention, convolution, LSTM, GRU (and generic RNN), and embedding, right out of the box and provides the means for supporting other user-defined layers. Opacus computes batched per-sample gradients, providing higher efficiency compared to the traditional "micro batch" approach. In this paper we present Opacus, detail the principles that drove its implementation and unique features, and benchmark it against other frameworks for training models with differential privacy as well as standard PyTorch.
Forward citations
Cited by 27 Pith papers
-
Dithered Gaussian Mechanism for Randomness-Efficient Differential Privacy
The dithered Gaussian mechanism discretizes the output of the Gaussian mechanism via a randomly shifted grid, inheriting Gaussian privacy guarantees while reducing private randomness to a constant per coordinate.
-
Reducing Per-Sample Harm in Stochastic Optimization
A batch-space trust-region step that reweights samples by last-layer gradient structure yields small but consistent accuracy gains for SGD and AdamW on image classification.
-
Composition for Pufferfish Privacy
Pufferfish mechanisms compose linearly only under DP-style inequalities; any per-entry DP mechanism can be translated to a composable Pufferfish mechanism using the a(b)-influence curve.
-
Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
Using Hessian eigenvalues from public data to design correlated noise for DP-SGD improves accuracy by 1–4% over current DP-MF methods.
-
Private Hyperparameter Tuning with Ex-Post Guarantee
A random-dropping mechanism tunes hyperparameters under ex-post DP with about a 2x privacy blowup for the winning candidate, and extends to Rényi DP.
-
General-Purpose $f$-DP Estimation and Auditing in a Black-Box Setting
A black-box estimator and auditor recover an algorithm's entire f-differential privacy trade-off curve with formal convergence and confidence guarantees.
-
End-to-End Differential Privacy in Training Deep Neural Network Classifiers
Perturbing softmax outputs with the Dirichlet mechanism during training yields input-private, label-public classifiers that beat prior differentially private training accuracy on five image benchmarks.
-
An Interactive Framework for Finding the Optimal Trade-off in Differential Privacy
Modeling the privacy-accuracy Pareto front with sigmoid curves and letting users pick points on hypothetical fronts finds the optimal DP privacy budget with fewer queries and evaluations than pairwise baselines.
-
Improving Noise Efficiency in Privacy-preserving Dataset Distillation
Dosser improves differentially private dataset distillation by decoupling sampling from optimization and projecting signals into a learned subspace.
-
What is the Cost of Differential Privacy for Deep Learning-Based Trajectory Generation?
DP-SGD causes large utility loss in deep trajectory generation, a new DP mechanism for conditional inputs helps stabilize GANs, and GANs overtake diffusion models when formal privacy is required.
-
Training Dynamics Underlying Language Model Scaling Laws: Loss Deceleration and Zero-Sum Learning
Loss deceleration, a piecewise-linear break in log-log loss curves, is attributed to zero-sum learning where per-example gradients oppose one another, and scaling helps by mitigating it.
-
The Gaussian Mixing Mechanism: Renyi Differential Privacy via Gaussian Sketches
Gaussian sketching with additive Gaussian noise satisfies a closed-form Rényi differential privacy bound that is tighter than prior analyses and improves private linear and logistic regression.
-
Multi-level Certified Defense Against Poisoning Attacks in Offline Reinforcement Learning
A DP-based certified defense provides lower bounds on expected cumulative reward and per-state action stability for offline RL under transition- and trajectory-level poisoning, with larger certified radii than COPA.
-
Hyperparameters in Score-Based Membership Inference Attacks
A new shadow-model hyperparameter selection method (KL-LiRA) makes membership inference attacks nearly as effective without knowing target hyperparameters, and training-data-based hyperparameter tuning shows no detect...
-
Balancing Utility and Privacy: Dynamically Private SGD with Random Projection
D2P2-SGD combines time-decreasing privacy noise with random projection to improve the accuracy of differentially private SGD, with convergence rates matching ordinary SGD.
-
Achieving Hilbert-Schmidt Independence Under R\'enyi Differential Privacy for Fair and Private Data Generation
FLIP combines a VAE, latent diffusion, Rényi DP, and CKA alignment across protected groups to produce tabular data with substantially reduced predictability of the protected attribute.
-
Uncertainty-Driven Reliability: Selective Prediction and Trustworthy Deployment in Modern Machine Learning
A training-dynamics abstention method matches deep ensembles at a fraction of the training cost, and a five-term error budget explains why selective classifiers still fall short of the oracle.
-
Embedding-Based Federated Data Sharing via Differentially Private Conditional VAEs
A federated, differentially private conditional VAE trained on foundation-model embeddings lets hospitals share synthetic data that supports downstream classification better than standard federated classifiers.
-
FlashDP: Private Training Large Language Models with Efficient DP-SGD
FlashDP fuses per-sample gradient computation, norm calculation, clipping, and noise addition into a cache-friendly block-wise all-reduce workflow that avoids explicit per-sample gradient storage and redundant recomputation.
-
Can One Safety Loop Guard Them All? Agentic Guard Rails for Federated Computing
A unified two-layer safety loop for federated computing that decouples guardrails from FHE, MPC, and DP back-ends via a backend-neutral DSL and interchangeable execution providers.
-
Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping
Bounded adaptive clipping puts a floor under the adaptive clipping threshold in DP-SGD and improves worst-class accuracy on skewed and Fashion MNIST.
-
A Robust Framework for Secure Cardiovascular Risk Prediction: An Architectural Case Study of Differentially Private Federated Learning
On synthetic cardiac data, FedCVR — a re-implementation of FedAdam with server-side momentum — is reported to reach F1 0.78 / AUC 0.96 under DP (ε≈13.4), beating stateless and other adaptive baselines, though the pape...
-
PrivacyBench: Privacy Isn't Free in Hybrid Privacy-Preserving Vision Systems
Combining federated learning with differential privacy causes catastrophic accuracy loss and large resource overhead in vision models, whereas federated learning with secure multi-party computation retains near-baseli...
-
Network-Aware Differential Privacy
Network-Aware Differential Privacy initiates a research agenda connecting networking and differential privacy, with preliminary evidence that packet-level adversaries can manipulate local DP protocols.
-
PPFL-RDSN: Privacy-Preserving Federated Learning-based Residual Dense Spatial Networks for Encrypted Lossy Image Reconstruction
A federated, privacy-preserving RDSN framework for encrypted image reconstruction whose local differential privacy mechanism is not actually differentially private because it releases low-frequency DCT coefficients wi...
-
Inclusive Federated Learning Through Compliance-Weighted Noise Allocation in Healthcare AI
Compliance-weighted noise allocation in federated healthcare learning claims no accuracy loss versus uniform noise, but its differential privacy guarantee applies only to the aggregator dataset, not client data.
-
How to DP-fy Your Data: A Practical Guide to Generating Synthetic Data With Differential Privacy
A practical, extremely thorough survey of differentially private synthetic data generation: methods, privacy units, evaluation metrics, and end-to-end system components across four data modalities.
Discussion (0). Continue with ORCID to comment.