I've got a dataset consisting of only 17 samples and 6 continuous features (all values in the dataset contain decimals, although 2 features exhibit categorical-ish behaviour). I'm looking at the possibility of employing GANs (or some other algorithm) to generate synthetic data points so that the risk of overfitting in my regression machine learning models is eliminated. Sadly it's impossible for me to get any more data points.

Do you people know about any generative algorithms that could successfully generate quality data from such an incredibly small dataset? I came across a paper titled "GenerativeMTD: A deep synthetic data generation framework for small datasets", however the smallest regression dataset for which the authors generate synthetic data is composed of 48 data points.

Thank you in advance!



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.