626_privacy/tensorflow_privacy
2
0
Fork 1
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions

Adding no-privacy baseline (dpsgd=False) to README.md.

Browse source 
PiperOrigin-RevId: 231004783
This commit is contained in:
Ilya Mironov 2019-01-25 18:18:42 -08:00 committed by schien1729
parent 668888c1a6
commit ff295b6a70
1 changed files with 6 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
tutorials/README.md
View file

@ -86,11 +86,13 @@ parameters satisfies differential privacy with eps = 2.92 and delta = 1e-05.
## Select Parameters ## Select Parameters
The table below has a few sample parameters illustrating various accuracy/privacy The table below has a few sample parameters illustrating various accuracy/privacy
tradeoffs (the first line is the default setting; privacy epsilon is reported tradeoffs (default parameters are in __bold__; privacy epsilon is reported
at delta=1e-5; accuracy is averaged over 10 runs). at delta=1e-5; accuracy is averaged over 10 runs).
| Learning rate | Noise multiplier | Clipping threshold | Number of microbatches | Number of epochs | Privacy eps | Accuracy | | Learning rate | Noise multiplier | Clipping threshold | Number of microbatches | Number of epochs | Privacy eps | Accuracy |
| ------------- | ---------------- | ----------------- | --------------------- | ---------------- | ----------- | -------- | | ------------- | ---------------- | ----------------- | --------------------- | ---------------- | ----------- | -------- |
| 0.08 | 1.12 | 1.0 | 256 | 60 | 2.92 | 96.6% | | 0.1 | | | __256__ | 10 | no privacy | 98.8% |
| 0.4 | 0.6 | 1.0 | 256 | 30 | 9.74 | 97.3% | | 0.32 | 1.2 | __1.0__ | __256__ | 10 | 1.20 | 95.0% |
| 0.32 | 1.2 | 1.0 | 256 | 10 | 1.20 | 95.0% | | __0.08__ | __1.12__ | __1.0__ | __256__ | __60__ | 2.92 | 96.6% |
| 0.4 | 0.6 | __1.0__ | __256__ | 30 | 9.74 | 97.3% |