LAVASET is an enhanced Random Forest method that can address correlation in datasets by grouping features based on different correlation metrics and creating fully interpretable latent variables; it reduces noise in feature importance and assigns more accurate importance values
Excited to share with everyone our new ML method, LAVASET, developed by @JmPosma and myself!
Special thanks to Kexin Xu, @DrDeclanORegan, @drjamesware, and Tim Ebbels for their valuable input and data sharing. https://t.co/WUrTk81OpW
CONFIRMED: Analysis of data obtained over the past 2 weeks by the #DARTMission team shows impact with Dimorphos has successfully altered the asteroid’s orbit by 32 minutes - marking the 1st time humans have changed the orbit of a celestial object in space! https://t.co/MjmUAFwVSO