2 ms·

I'm sorry, but if you're generating the inputs to your model based on a model, you've invented a pathological chicken and egg problem. I suppose as long as your

I'm sorry, but if you're generating the inputs to your model based on a model, you've invented a pathological chicken and egg problem. I suppose as long as your model never changes this will work well, but what happens when you want to change your model, but already have stored synthetic data? Tangent: I suppose humans choice in models and statistics to use can be argued to be a similar thing, but I'm still dubious of how this concept is being communicated.

lettergram 7y ago
Take a dataset, extract the features, send to model, design model, check if we match desired output. That’s the standard Data science workflow. What we do is extract features in an unsupervised way and build a model to recreate the dataset from said feature. When the synthesized dataset is indistinguishable from the real thing. You should be able to build a model that gets you 90%-100% of the way there with synthetic data, then tune / retrain on the real data. The model that does decisioning itself doesn’t matter. If you needed to adjust the synthetic dataset, you’d have to either build a new generative model OR otherwise weight the dataset.