Celestial Machine Learning: From Data to Mars and Beyond with AI Feynman

This work addresses the challenge of automated scientific discovery from data for astronomers and physicists, but it is incremental as it applies an existing method to a new dataset.

The paper tackled the problem of discovering Kepler's first law from astronomical data alone, using AI Feynman for symbolic regression to emulate Kepler's discovery with the Rudolphine tables, achieving successful emulation of the orbital equation.