Finding specific datasets to train and test your models on can be challenging depending on the task at hand. Data related to loan applications is especially problematic as there are very few public datasets in existence. This challenge is about generating a synthetic dataset that can then be used to test a variety of algorithms.

 

Creating this synthetic dataset will be achieved by collating a variety of open source data and likely using Generative Adversarial Networks among other techniques.

 

Can we generate realistic synthetic datasets to help credit risk estimation? Help this team find out.

This team is looking for 1-2 data scientists and 1-2 machine learning engineers.