Simple decision trees and complex image generators are actually the exact same thing mathematically.
Decision trees split data into categories like a flowchart while diffusion models generate detailed images from random noise. A shared mathematical principle called Global Trajectory Score Matching links these discrete and continuous processes. This unification means that techniques used to speed up simple trees can now be applied to massive generative models. It bridges the gap between old-school interpretable machine learning and modern generative systems. Developers can leverage this link to build generative AI that is much easier to inspect and understand.
Trees to Flows and Back: Unifying Decision Trees and Diffusion Models
arXiv · 2605.00414
Decision trees and diffusion models are ostensibly disparate model classes, one discrete and hierarchical, the other continuous and dynamic. This work unifies the two by establishing a crisp mathematical correspondence between hierarchical decision trees and diffusion processes in appropriate limiting regimes. Our unification reveals a shared optimization principle: \emph{Global Trajectory Score Matching (GTSM)}, for which gradient boosting (in an idealized version) is asymptotically optimal. We