-
### Describe the workflow you want to enable
Add gradient clipping to SGDRegressor to:
1. Improve training stability when dealing with outliers or ill-conditioned data
2. Enable differentially priv…
-
### Describe the workflow you want to enable
Enable differential privacy in SGDRegressor by adding noise injection with:
1. Manual noise scale setting, or
2. Automatic noise calibration from de…
-
Differential Privacy is implemented but never used. It seems that it was intentionally removed during a code refactoring (in [this PR](https://github.com/epfml/disco/pull/578/files#diff-bb5fd2b79ca040…
-
### Session description
@bmcase and I, along with several differential privacy researchers, have developed a compelling privacy framework where each device tracks and controls the privacy loss incurr…
-
## Issue Description
Hello everyone, I hope you're doing well. I'm trying to implement Differential Privacy (DP) with Opacus in a federated learning training structure where, in each round, each cl…
-
Differential Privacy is an important element of many FL workloads. The purpose of this issue is to track interest related to it. Please upvote this issue if you need a Flower-provided Differential Pri…
-
**Request**:
We propose allowing contributions of float values (including negative ones) to `contributeToHistogramOnEvent` with a [sensitivity](https://en.wikipedia.org/wiki/Additive_noise_differen…
-
Hi, I would like to know where the computation of privacy budget is in differential privacy?
-
## Overview
Investigate and implement Dataverse support for sensitive data using DataTags and Differential Privacy.
## Tasks
### Background
- [x] Identify the sensitive data tools that we plan…
-
Hi,
https://github.com/google/differential-privacy/commit/af761ba5bc7fb0007ed8bdde62db77046d1aff8b added "MixtureGaussianPrivacyLoss" to `dp_accounting/pld/privacy_loss_mechanism.py`, which enables p…