Learning Functions of Halfspaces.
J. Alman and S. Patel and R. Servedio.
In 58th Annual Symposium on Theory of Computing (STOC), 2026.


Abstract:

We give an algorithm that learns arbitrary Boolean functions of $k$ arbitrary halfspaces over $\R^n$, in the challenging distribution-free Probably Approximately Correct (PAC) learning model, running in time $2^{\sqrt{n} \cdot (\log n)^{O(k)}}.$ This is the first algorithm that can PAC learn even intersections of two halfspaces in time $2^{o(n)}.$

Link to full version


Back to main papers page