This paper of Tseitin is a landmark as the first to give non-trivial lower bounds for propositional proofs; although it pre-dates the first papers on

In his seminal paper on probabilistic Turing machines, Gill [13] asked whether the class PP is closed under intersection and union. We give a positive answer to this question. We also show that PP is closed under a variety of polynomial-time truth-table reductions. Consequences in complexity theory include the definite collapse and (assuming P<F NaN> 6= PP) separation of certain query hierarchies over PP. Similar techniques allow us to combine several threshold gates into a single threshold gate. Consequences in the study of circuits include the simulation of circuits with a small number of threshold gates by circuits having only a single threshold gate at the root (perceptrons), and a lower bound on the number of threshold gates needed to compute the parity function. 1. Introduction The class PP was defined in 1972 by John Gill [13, 14] and independently by Janos Simon [26] in 1974. PP is the class of languages accepted by a polynomial-time bounded nondeterministic Turing machine t...

The recent approaches of extending the GRAPHPLAN algorithm to handle more expressive planning formalisms raise the question of what the formal meaning of “expressive power ” is. We formalize the intuition that expressive power is a measure of how concisely planning domains and plans can be expressed in a particular formalism by introducing the notion of “compilation schemes ” between planning formalisms. Using this notion, we analyze the expressiveness of a large family of propositional planning formalisms, ranging from basic STRIPS to a formalism with conditional effects, partial state specifications, and propositional formulae in the preconditions. One of the results is that conditional effects cannot be compiled away if plan size should grow only linearly but can be compiled away if we allow for polynomial growth of the resulting plans. This result confirms that the recently proposed extensions to the GRAPHPLAN algorithm concerning conditional effects are optimal with respect to the “compilability ” framework. Another result is that general propositional formulae cannot be compiled into conditional effects if the plan size should be preserved linearly. This implies that allowing general propositional formulae in preconditions and effect conditions adds another level of difficulty in generating a plan.

We show that for most complexity classes of interest, all sets complete under first-order projections are isomorphic under first-order isomorphisms. That is, a very restricted version of the Berman-Hartmanis Conjecture holds.

We prove new results on the circuit complexity of Approximate Majority, which is the problem of computing Majority of a given bit string whose fraction of 1’s is bounded away from 1/2 (by a constant). We then apply these results to obtain new relationships between probabilistic time, BPTime (t), and alternating time, Σ O(1)Time (t). Our main results are the following: 1. We prove that 2 n0.1�-size depth-3 circuits for Approximate Majority on n bits have bottom fan-in Ω(log n). As a corollary we obtain that BPTime (t) �⊆ Σ2Time � o(t 2) � with respect to some oracle. This complements the result that BPTime (t) ⊆ Σ2Time � t 2 · poly log t � with respect to every oracle (Sipser and Gács, STOC ’83; Lautemann, IPL ’83). 2. We prove that Approximate Majority is computable by uniform polynomial-size circuits of depth 3. Prior to our work, the only known polynomial-size depth-3 circuits for Approximate Majority were non-uniform (Ajtai, Ann. Pure Appl. Logic ’83). We also prove that BPTime (t) ⊆ Σ3Time (t · poly log t). This complements our results in (1). 3. We prove new lower bounds for solving QSAT 3 ∈ Σ3Time (n · poly log n) on probabilistic computational models. In particular, we prove that solving QSAT 3 requires time n 1+Ω(1) on Turing machines with a random-access input tape and a sequential-access work tape that is initialized with random bits. No lower bound was previously known on this model (for a function computable in linear space). ∗ Author supported by NSF grant CCR-0324906. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the

Several recent results in circuit complexity theory have used a representation of Boolean functions by polynomials over finite fields. Our current inability to extend these results to superficially similar situations may be related to properties of these polynomials which do not extend to polynomials over general finite rings or finite abelian groups. Here we pose a number of conjectures on the behavior of such polynomials over rings and groups, and present some partial results toward proving them. 1. Introduction 1.1. Polynomials and Circuit Complexity The representation of Boolean functions as polynomials over the finite field Z 2 = f0; 1g dates back to early work in switching theory [?]. A formal language L can be identified with the family of functions f i : Z i 2 ! Z 2 , where f i (x 1 ; : : : ; x i ) = 1 iff x 1 : : : x i 2 L. Each of these functions can be written as a polynomial in the variables x 1 ; : : : ; x n . We can consider algebraic formulas or circuits with...

We construct a randomness-efficient averaging sampler that is computable by uniform constant-depth circuits with parity gates (i.e., in uniform AC 0 [⊕]). Our sampler matches the parameters achieved by random walks on constant-degree expander graphs, allowing us to apply a variety expander-based techniques within NC 1. For example, we obtain the following results:

We show that there are functions computable by linear size boolean circuits of depth k that require super-polynomial size perceptrons of depth k \Gamma 1, for k ! logn=(6loglogn). This result implies the existence of an oracle A such that S p;A k 6` PP S p;A k\Gamma2 and in particular this oracle separates the levels in the PP PH hierarchy. Using the same ideas, we show a lower bound for another function, which makes it possible to strengthen the oracle separation to D p;A k 6` PP S p;A k\Gamma2 . 1 Introduction There is a strong connection between lower bounds for boolean circuits (consisting of AND, OR, and NOT gates) and relativization results about the polynomial time hierarchy. This fact was first established by Furst, Saxe, and Sipser [5]. Sipser [13] later defined a family of functions that are computable by linear size circuits of depth k, and showed that they require super-polynomial size boolean circuits of depth k \Gamma 1. Yao [14] and Hstad [8, 9] impr...

We study the task of transforming a hard function f, with which any small circuit disagrees on (1 − δ)/2 fraction of the input, into a harder function f ′ , with which any small circuit disagrees on (1 − δ k)/2 fraction of the input, for δ ∈ (0, 1) and k ∈ N. We show that this process can not be carried out in a black-box way by a circuit of depth d and size 2 o(k1/d) or by a nondeterministic circuit of size o(k / log k) (and arbitrary depth). In particular, for k = 2 Ω(n) , such hardness amplification can not be done in ATIME(O(1), 2 o(n)). Therefore, hardness amplification in general requires a high complexity. Furthermore, we show that even without any restriction on the complexity of the amplification procedure, such a black-box hardness amplification must be inherently non-uniform in the following sense. Given as an oracle any algorithm which agrees with f ′ on (1 − δ k)/2 fraction of the input, we still need an additional advice of length Ω(k log(1/δ)) in order to compute f correctly on (1−δ)/2 fraction of the input. Therefore, to guarantee the hardness, even against uniform machines, of the function f ′ , one has to start with a function f which is hard against non-uniform circuits. Finally, we derive similar lower bounds for any black-box construction of pseudorandom generators from hard functions.

Program checking, program self-correcting and program self-testing were pioneered by [Blum and Kannan] and [Blum, Luby and Rubinfeld] in the mid eighties as a new way to gain confidence in software, by considering program correctness on an input by input basis rather than full program verification. Work in the field of program checking focused on designing, for specific functions, checkers, testers and correctors that are more efficient than the best program known for the function. These were designed utilizing specific algebraic, combinatorial or completeness properties of the function at hand. In this work we introduce a novel composition methodology for improving the efficiency of program checkers. We use this approach to design a variety of program checkers that are provably more efficient, in terms of circuit depth, than the optimal program for computing the function being checked. Extensions of this methodology for the cases of program testers and correctors are also presented. In particular, we show: • For all i ≥ 1, every language in RNC i (that is NC 1-hard under NC 0-reductions) has a program checker in RNC i−1.