On (not) learning the Möbius function
For researchers in learning theory and computational number theory, it establishes fundamental limitations of standard learning techniques for a specific number-theoretic function.
The paper proves lower bounds on learning the Möbius function using kernel methods, noisy gradient methods, and correlational statistical query algorithms, based on quantitative correlation bounds with digital characters.
We prove lower bounds on learning the Möbius or Liouville function with a variety of standard learning techniques, including kernel methods, noisy gradient methods, and correlational statistical query algorithms. These results follow from quantitative bounds on the correlation of Möbius with digital characters of various finite abelian groups, where the group is dictated by the type of input data the algorithm is given. Using residues mod $p$ for many different primes corresponds to a cyclic group, and using the base $p$ expansion for a fixed prime corresponds to an elementary abelian $p$-group. We also note that lower bounds of this form are closely related to certain types of digital prime number theorems.