Abstract. For every [Formula: see text], we construct a sum-of-squares identity [Formula: see text], where [Formula: see text] are bilinear forms with complex coefficients and [Formula: see text]. Previously, such a construction was known with [Formula: see text]. The same bound holds over any field of positive characteristic. As an application to complexity of noncommutative computation, we show that the polynomial [Formula: see text] in [Formula: see text] noncommuting variables can be computed by a noncommutative arithmetic circuit of size [Formula: see text]. This holds over any field of characteristic different from two. The same bound applies to noncommutative versions of the elementary symmetric polynomial of degree four and the rectangular permanent of a [Formula: see text] matrix.

We consider read-k determinantal representations of polynomials and prove some non-expressibility results. A square matrix M whose entries are variables or field elements will be called read-k, if every variable occurs at most k times in M. It will be called a determinantal representation of a polynomial f if f = det(M). We show that - the n × n permanent polynomial does not have a read-k determinantal representation for k ∈ o(√n/log n) (over a field of characteristic different from two). We also obtain a quantitative strengthening of this result by giving a similar non-expressibility for k ∈ o(√n/log n) for an explicit n-variate multilinear polynomial (as opposed to the permanent which is n²-variate).

Abstract Given a non-negative real matrix M of non-negative rank at least r , can we witness this fact by a small submatrix of M ? While Moitra (SIAM J. Comput. 2013) proved that this cannot be achieved exactly, we show that such a witnessing is possible approximately: An $$m\times n$$ m × n matrix of non-negative rank r always contains a submatrix with at most r 3 rows and columns with non-negative rank at least $$\Omega(\frac{r}{\log n\log m})$$ Ω ( r log n log m ) . A similar result is proved for the 1-partition number of a Boolean matrix and, consequently, also for its two-player deterministic communication complexity. Tightness of the latter estimate is closely related to the log-rank conjecture of Lovász and Saks.

Given a non-negative real matrix M of non-negative rank at least r, can we witness this fact by a small submatrix of M? While Moitra (SIAM J. Comput. 2013) proved that this cannot be achieved exactly, we show that such a witnessing is possible approximately: an m×n matrix of non-negative rank r always contains a submatrix with at most r³ rows and columns with non-negative rank at least Ω(r/(log n log m)). A similar result is proved for the 1-partition number of a Boolean matrix and, consequently, also for its two-player deterministic communication complexity. Tightness of the latter estimate is closely related to the log-rank conjecture of Lovász and Saks.

We give several new lower bounds on size of homogeneous non-commutative circuits. We present an explicit homogeneous bivariate polynomial of degree $d$ which requires homogeneous non-commutative circuit of size $Ω(d/\log d)$. For an $n$-variate polynomial with $n>1$, the result can be improved to $Ω(nd)$, if $d\leq n$, or $Ω(nd \frac{\log n}{\log d})$, if $d\geq n$. Under the same assumptions, we also give a quadratic lower bound for the ordered version of the central symmetric polynomial.

We define the shadow complexity of a polytope P as the maximum number of vertices in a linear projection of P to the plane. We describe connections to algebraic complexity and to parametrized optimization. We also provide several basic examples and constructions, and develop tools for bounding shadow complexity.

A fundamental result in statistical learning theory is the equivalence of PAC learnability of a class with the finiteness of its Vapnik-Chervonenkis dimension. However, this clean result applies only to binary classification problems. In search for a similar combinatorial characterization of learnability in a more general setting, we discovered a surprising independence of set theory for some basic general notion of learnability. Consider the following statistical estimation problem: given a family F of real valued random variables over some domain X and an i.i.d. sample drawn from an unknown distribution P over X, find f in F such that its expectation w.r.t. P is close to the supremum expectation over all members of F. This Expectation Maximization (EMX) problem captures many well studied learning problems. Surprisingly, we show that the EMX learnability of some simple classes depends on the cardinality of the continuum and is therefore independent of the set theory ZFC axioms. Our results imply that that there exist no "finitary" combinatorial parameter that characterizes EMX learnability in a way similar to the VC-dimension characterization of binary classification learnability.

There are various notions of balancing set families that appear in combinatorics and computer science. For example, a family of proper non-empty subsets S_1,...,S_k subset [n] is balancing if for every subset X subset {1,2,...,n} of size n/2, there is an i in [k] so that |S_i cap X| = |S_i|/2. We extend and simplify the framework developed by Hegedűs for proving lower bounds on the size of balancing set families. We prove that if n=2p for a prime p, then k >= p. For arbitrary values of n, we show that k >= n/2 - o(n). We then exploit the connection between balancing families and depth-2 threshold circuits. This connection helps resolve a question raised by Kulikov and Podolskii on the fan-in of depth-2 majority circuits computing the majority function on n bits. We show that any depth-2 threshold circuit that computes the majority on n bits has at least one gate with fan-in at least n/2 - o(n). We also prove a sharp lower bound on the fan-in of depth-2 threshold circuits computing a specific weighted threshold function.

We prove new lower bounds on the sizes of proofs in the Cutting Plane proof system, using a concept that we call unsatisfiability certificate. This approach is, essentially, equivalent to the well-known feasible interpolation method, but is applicable to CNF formulas that do not seem suitable for interpolation. Specifically, we prove exponential lower bounds for random k-CNFs, where k is the logarithm of the number of variables, and for the Weak Bit Pigeon Hole Principle. Furthermore, we prove a monotone variant of a hypothesis of Feige [12]. We give a superpolynomial lower bound on monotone real circuits that approximately decide the satisfiability of k-CNFs, where k = ω(1). For k ≈ log n, the lower bound is exponential.

The isoperimetric profile of a graph is a function that measures, for an integer k, the size of the smallest edge boundary over all sets of vertices of size k. We observe a connection between isoperimetric profiles and computational complexity. We illustrate this connection by an example from communication complexity, but our main result is in algebraic complexity. We prove a sharp super-polynomial separation between monotone arithmetic circuits and monotone arithmetic branching programs. This shows that the classical simulation of arithmetic circuits by arithmetic branching programs by Valiant, Skyum, Berkowitz, and Rackoff (1983) cannot be improved, as long as it preserves monotonicity. A key ingredient in the proof is an accurate analysis of the isoperimetric profile of finite full binary trees. We show that the isoperimetric profile of a full binary tree constantly fluctuates between one and almost the depth of the tree.

In this paper, we compare the strength of the semantic and syntactic version of the cutting planes proof system. First, we show that the lower bound technique of Pudlák applies also to semantic cutting planes: the proof system has feasible interpolation via monotone real circuits, which gives an exponential lower bound on lengths of semantic cutting planes refutations. Second, we show that semantic refutations are stronger than syntactic ones. In particular, we give a formula for which any refutation in syntactic cutting planes requires exponential length, while there is a polynomial length refutation in semantic cutting planes. In other words, syntactic cutting planes does not p-simulate semantic cutting planes. We also give two incompatible integer inequalities which require exponential length refutation in syntactic cutting planes. Finally, we pose the following problem, which arises in connection with semantic inference of arity larger than two: can every multivariate non-decreasing real function be expressed as a composition of non-decreasing real functions in two variables?

We consider boolean circuits in which every gate may compute an arbitrary boolean function of k other gates, for a parameter k. We give an explicit function $f:{0,1}^n -> {0,1} that requires at least Omega(log^2(n)) non-input gates when k = 2n/3. When the circuit is restricted to being layered and depth 2, we prove a lower bound of n^(Omega(1)) on the number of non-input gates. When the circuit is a formula with gates of fan-in k, we give a lower bound Omega(n^2/k*log(n)) on the total number of gates. Our model is connected to some well known approaches to proving lower bounds in complexity theory. Optimal lower bounds for the Number-On-Forehead model in communication complexity, or for bounded depth circuits in AC_0, or extractors for varieties over small fields would imply strong lower bounds in our model. On the other hand, new lower bounds for our model would prove new time-space tradeoffs for branching programs and impossibility results for (fan-in 2) circuits with linear size and logarithmic depth. In particular, our lower bound gives a different proof for a known time-space tradeoff for oblivious branching programs.

We study arithmetic proof systems ${\mathbb P}_c({\mathbb F})$ and $ {\mathbb P}_f({\mathbb F})$ operating with arithmetic circuits and arithmetic formulas, respectively, and that prove polynomial identities over a field ${\mathbb F}$. We establish a series of structural theorems about these proof systems, the main one stating that ${\mathbb P}_c({\mathbb F})$ proofs can be balanced: if a polynomial identity of syntactic degree $ d $ and depth $k$ has a ${\mathbb P}_c({\mathbb F})$ proof of size $s$, then it also has a ${\mathbb P}_c({\mathbb F})$ proof of size $ {\rm poly}(s,d) $ in which every circuit has depth $ O(k+\log^2 d + \log d\cdot \log s) $. As a corollary, we obtain a quasi-polynomial simulation of ${\mathbb P}_c({\mathbb F})$ by ${\mathbb P}_f({\mathbb F})$. Using these results we obtain the following: consider the identities $\det(XY) = \det(X)\cdot\det(Y) \mbox{ and } \det(Z)= z_{11}\cdots z_{nn},$ where $X,Y$, and $ Z$ are $n\times n$ square matrices and $Z$ is a triangular matrix with $z_{11},\dots, z_{nn}$ on the diagonal (and $ \det $ is the determinant polynomial). Then we can construct a polynomial-size arithmetic circuit $\det$ such that the above identities have ${\mathbb P}_c({\mathbb F})$ proofs of polynomial size using circuits of $ O(\log^2 n)$ depth. Moreover, there exists an arithmetic formula $ \det $ of size $n^{O(\log n)}$ such that the above identities have ${\mathbb P}_f({\mathbb F})$ proofs of size $n^{O(\log n)}$. This yields a solution to a basic open problem in propositional proof complexity, namely, whether there are polynomial-size $\mathbf{NC}^2$-Frege proofs for the determinant identities and the hard matrix identities, as considered, e.g., in Soltys and Cook [Ann. Pure Appl. Logic, 130 (2004), pp. 277--323] (cf. Beame and Pitassi [Bull. Eur. Assoc. Theor. Comput. Sci. EATCS, 65 (1998), pp. 66--89]). We show that matrix identities like $ AB=I \rightarrow BA=I $ (for matrices over the two element field) as well as basic properties of the determinant have polynomial-size $\mathbf{NC}^2$-Frege proofs and quasi-polynomial-size Frege proofs.

We initiate the study of the complexity of arithmetic circuits with division gates over non-commuting variables. Such circuits and formulas compute non-commutative rational functions, which, despite their name, can no longer be expressed as ratios of polynomials. We prove some lower and upper bounds, completeness and simulation results, as follows.

We give an example of a non-commutative mono-tone polynomial f which can be computed by a polynomial-size non-commutative formula, but every monotone non-commutative circuit computing f must have an exponential size. In the non-commutative setting this gives, a fortiori, an exponential separation between monotone and general formulas, monotone and general branching programs, and monotone and general circuits. This answers some questions raised by Nisan.

We study arithmetic proof systems Pc(F) and Pf(F) operating with arithmetic circuits and arithmetic formulas, respectively, that prove polynomial identities over a field F. We establish a series of structural theorems about these proof systems, the main one stating that Pc(F) proofs can be balanced: if a polynomial identity of syntactic degree d and depth k has a Pc(F) proof of size s, then it also has a Pc(F) proof of size poly(s,d) and depth O(k+log2 d + log d• log s). As a corollary, we obtain a quasipolynomial simulation of Pc(F) by Pf(F), for identities of a polynomial syntactic degree. Using these results we obtain the following: consider the identities: det(XY) = det(X)•det(Y) and det(Z)= z11 ••• znn, where X,Y and Z are n x n square matrices and Z is a triangular matrix with z11,..., znn on the diagonal (and det is the determinant polynomial). Then we can construct a polynomial-size arithmetic circuit det such that the above identities have Pc(F) proofs of polynomial-size and O(log2n) depth. Moreover, there exists an arithmetic formula det of size nO(log n) such that the above identities have Pf(F) proofs of size nO(log n).

This paper extends Valiant's work on VP and VNP to the settings in which variables are not multiplicatively commutative and/or associative. Our main result is a theory of completeness for these algebraic worlds. We define analogs of Valiant's classes VP and VNP, as well as of the polynomials permanent and determinant, in these worlds. We then prove that even in a completely relationless world which assumes no commutativity nor associativity, permanent remains VNP-complete, and determinant can polynomially simulate any arithmetic formula, just as in the standard commutative, associative world of Valiant. In the absence of associativity, the completeness proof gives rise to the following combinatorial problem: what is the smallest binary tree which contains as minors all binary trees with n leaves. We give an explicit construction of such a universal tree of polynomial size, a result of possibly independent interest. Given that such non-trivial reductions are possible even without commutativity and associativity, we turn to lower bounds. In the non-associative, commutative world we prove exponential circuit lower bounds on explicit polynomials, separating the non-associative commutative analogs of VP and VNP. Obtaining such lower bounds and a separation in the complementary associative, non-commutative world has been open for about 30 years.

We initiate a direction for proving lower bounds on the size of non-commutative arithmetic circuits. This direction is based on a connection between lower bounds on the size of non-commutative arithmetic circuits and a problem about commutative degree four polynomials, the classical sum-of-squares problem: find the smallest n such that there exists an identity (x12+x22+•• + xk2)• (y1^2+y22+•• + yk2)= f12+f22+ ... +fn2, where each fi = fi(X,Y) is bilinear in X={x1,... ,xk} and Y={y1,..., yk}. Over the complex numbers, we show that a sufficiently strong super-linear lower bound on n in, namely, n ≥ k1+ε with ε >0, implies an exponential lower bound on the size of arithmetic circuits computing the non-commutative permanent.

Devising an efficient deterministic - or even a non-deterministic sub-exponential time - algorithm for testing polynomial identities is a fundamental problem in algebraic complexity and complexity at large. Motivated by this problem, as well as by results from proof complexity, we investigate the complexity of proving polynomial identities. To this end, we study a class of equational proof systems, of varying strength, operating with polynomial identities written as arithmetic formulas over a given ring. A proof in these systems establishes that two arithmetic formulas compute the same polynomial, and consists of a sequence of equations between polynomials, written as arithmetic formulas, where each equation in the sequence is derived from previous equations by means of the polynomial-ring axioms. We establish the first non-trivial upper and lower bounds on the size of equational proofs of polynomial identities, as follows: 1. Polynomial-size upper bounds on equational proofs of identities involving symmetric polynomials and interpolation-based identities. In particular, we show that basic properties of the elementary symmetric polynomials are efficiently provable already in equational proofs operating with depth-4 formulas, over infinite fields. This also yields polynomial-size depth-4 proofs of the Newton identities, providing a positive answer to a question posed by Grigoriev and Hirsch. 2. Exponential-size lower bounds on (full, unrestricted) equational proofs of identities over certain specific rings. 3. Exponential-size lower bounds on analytic proofs operating with depth-3 formulas, under a certain regularity condition. The "analytic" requirement is, roughly, a condition that forbids introducing arbitrary formulas in a proof and the regularity condition is an additional structural restriction. 4. Exponential-size lower bounds on one-way proofs (of unrestricted depth) over infinite fields. Here, one-way proofs are analytic proofs, in which one is also not allowed to introduce arbitrary constants. Furthermore, we determine basic structural characterizations of equational proofs, and consider relations with polynomial identity testing procedures. Specifically, we show that equational proofs efficiently simulate the polynomial identity testing algorithm provided by Dvir and Shpilka.

The paper presents a model of the emission load in the vicinity of a monitored road in relation to the real traffic input data. It describes a simple method of the emission load estimating. In Addition to the modeling method itself, the paper describes particular methodologies of the real data conversion and processing. Discussed are traffic intensities of the heavy trucks over 12 tons of weight. The model uses as its input real traffic data files from intelligent traffic systems (ITS) localized by the monitored roads. More specifically, data from selected highway toll gates were used in this work. At the end, particular and final, results are presented in graphs as examples.

Abstract We prove that Kreisel's Conjecture is true, if Peano arithmetic is axiomatised using minimality principle and axioms of identity (theory PAM). The result is independent on the choice of language of PAM. We also show that if infinitely many instances of A(x) are provable in a bounded number of steps in PAM then there exists . The results imply that PAM does not prove scheme of induction or identity schemes in a bounded number of steps.

Abstract We give four examples of theories in which Kreisel's Conjecture is false: (1) the theory PA(-) obtained by adding a function symbol minus, ‘—’, to the language of PA, and the axiom ∀x∀y∀z (x − y = z) ≡ (x = y + z ∨ (x < y ∧ z = 0)); (2) the theory L of integers; (3) the theory PA(q) obtained by adding a function symbol q (of arity ≥ 1) to PA, assuming nothing about q; (4) the theory PA(N) containing a unary predicate N(x) meaning ‘x is a natural number’. In Section 6 we suggest a counterexample to the so called Sharpened Kreisel's Conjecture.

Abstract We give an exponential lower bound on number of proof-lines in the proof system K of modal logic, i.e., we give an example of K-tautologies ψ1, ψ2, … s.t. every K-proof of ψi must have a number of proof-lines exponential in terms of the size of ψi. The result extends, for the same sequence of K-tautologies, to the systems K4, Gödel–Löb's logic, S andS4. We also determine some speed-up relations between different systems of modal logic on formulas of modal-depth one.