The project is being inspired by the fact that machine learning is at the forefront of innovation in the world we live in today and it is important for developers to be familiar with not only deep learning frameworks like pytorch to train dog and cat classifiers but also leverage libraries like pysyft to help tackle and train machine learning models to solve real world human problems like cancer, depression and anxiety in the healthcare sector.These problems are what developers should look forward to solving with deep learning but due to the private data involved in training these models, little or no technical knowledge on privacy preserving tools and also unavailability of opensource frameworks to facilitate training, developers tend to focus on working with toy dataset which are readily available for hand written igits recognition or dog and cat classifiers.

What it does

This project is a step by step tutorial guide that introduces pytorch users to an open source library pysyft which is an extension of pytorch for performing training on data you do not have access to and also introduces them to some other privacy preserving techniques.

How I built it

The project was developed with PyTorch and PySyft an open source tool for privacy preserving machine learning

Challenges I ran into

I had challenges understanding some concepts in the privacy preserving context like differential privacy, secure multiparty computation and homomorphic encryption

Accomplishments that I'm proud of

i am proud of the fact that i can leverage pytorch and pysyft in developing sydtems that preserve privacy of data.

What I learned

i learnt a couple of new machine learning algorithms alongside how to implement them in code.

Built With

Share this project: