Going from the parent comment's assertion, the answer would be that if you want to train something to predict physics, the original physics simulation that's generating the data is either:
1. Already doing that, in which case it's already the superior model.
or
2. Not doing that correctly, in which case the synthetic data output is equally poor.
Consider computer graphics and computer vision. If you had a photorealistic renderer, it could be useful for generating training data for a vision algorithm. However, you can't just run your rendering pipeline backwards to create a scene out of an image.
Yes. A typical example would be the use of statistical emulators, emulating process (physical) based models that are computationally complex in the environmental sciences.
1. Already doing that, in which case it's already the superior model.
or
2. Not doing that correctly, in which case the synthetic data output is equally poor.