A Separation Between Distribution-Free SQ Learning and Dimension Complexity

RESUMO

We show that there exists a class of boolean functions C such that $(i)$ there is a distribution-independent statistical query algorithm for learning C that makes a polynomial number of queries of inverse polynomial tolerance and $(ii)$ for any set of functions $\Phi_1, \dots, \Phi_r$ such that for all $f \in$ C we can write $f(x) = \text{sign} \left( \sum_{i = 1}^r w_i \Phi_i(x) \right)$ for some set of weights $w_i \in \mathbb{R}$, we must have that $r \geq n^{\omega(1)}$. This gives a superpolynomial separation between dimension complexity and the query complexity of distribution-free learning in the statistical query model, negatively answering a question of Feldman, Kamath, and Srebro [FKS26]. Our construction C is a subclass of DNFs, and the proof is a simple consequence of recent progress on agnostically learning conjunctions [DKR25,CPS26] and the work of Razborov and Sherstov on the sign rank of DNFs [RS10].