Improved Learning of AC0 Functions
نویسندگان
چکیده
Two extensions of the Linial, Mansour, Nisan AC0 learning algorithm are presented. The LMN method works when input examples are drawn uniformly. The new algorithms improve on theirs by performing well when given inputs drawn from unknown, mutually independent distributions. A variant of the one of the algorithms is conjectured to work in an even broader setting.
منابع مشابه
Agnostic Learning from Tolerant Natural Proofs
We generalize the “learning algorithms from natural properties” framework of [4] to get agnostic learning algorithms from natural properties with extra features. We show that if a natural property (in the sense of Razborov and Rudich [28]) is useful also against functions that are close to the class of “easy” functions, rather than just against “easy” functions, then it can be used to get an ag...
متن کاملImproved Bounds on the Sign-Rank of AC^0
The sign-rank of a matrix A with entries in {−1,+1} is the least rank of a real matrix B with Aij · Bij > 0 for all i, j. Razborov and Sherstov (2008) gave the first exponential lower bounds on the sign-rank of a function in AC0, answering an old question of Babai, Frankl, and Simon (1986). Specifically, they exhibited a matrix A = [F (x, y)]x,y for a specific function F : {−1, 1} × {−1, 1} → {...
متن کاملSeparation of AC0[⊕] Formulas and Circuits
This paper gives the first separation between the power of formulas and circuits of equal depth in the AC0[⊕] basis (unbounded fan-in AND, OR, NOT and MOD2 gates). We show, for all d(n) ≤ O( logn log logn ), that there exist polynomial-size depth-d circuits that are not equivalent to depth-d formulas of size no(d) (moreover, this is optimal in that no(d) cannot be improved to nO(d)). This resul...
متن کاملLearning Algorithms from Natural Proofs
Based on Håstad’s (1986) circuit lower bounds, Linial, Mansour, and Nisan (1993) gave a quasipolytime learning algorithm for AC0 (constant-depth circuits with AND, OR, and NOT gates), in the PAC model over the uniform distribution. It was an open question to get a learning algorithm (of any kind) for the class of AC0[p] circuits (constant-depth, with AND, OR, NOT, and MODp gates for a prime p)....
متن کاملOn Tc 0 , Ac 0 , and Arithmetic Circuits 1
Continuing a line of investigation that has studied the function classes #P [Val79b], #SAC1 [Val79a, Vin91, AJMV], #L [AJ93b, Vin91, AO94], and #NC1 [CMTV96], we study the class of functions #AC0. One way to de ne #AC0 is as the class of functions computed by constant-depth polynomial-size arithmetic circuits of unbounded fan-in addition and multiplication gates. In contrast to the preceding fu...
متن کامل