privacy preserving predictions

 
partydb_background_confetti.jpg
180613 PartyDB Illustration ideas-11.png

PartyDB runs private predictions on machine learning models. Models can be imported from TensorFlow, PyTorch, and other popular frameworks.

Predictions are securely computed, preserving the privacy of the inputs and outputs. PartyDB will make it easy for developers and data scientists to build a new category of privacy-preserving applications.

 
 

Secure Multi-Party Computation

Secure multi-party computation (mpc) enables us to run predictions on private input without revealing anything.

 
180613 PartyDB Illustration ideas-10.png
 
 

Secret Sharing

Secure multi-party computation (mpc) enables us to run predictions on private input without revealing anything.

 
 
 

ONNX Support

Secure multi-party computation (mpc) enables us to run predictions on private input without revealing anything.

180613 PartyDB Illustration ideas-05.png

demos

Install

paste Into Terminal:

/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
 

partydb installs packages to their own directory and then symlinks their files into /usr/local.

$ cd /usr/local
$ find Cellar
Cellar/wget/1.16.1
Cellar/wget/1.16.1/bin/wget
Cellar/wget/1.16.1/share/man/man1/wget.1

$ ls -l bin
bin/wget -> ../Cellar/wget/1.16.1/bin/wget

join the future of ai & ml