Regular Resolution Lower Bounds For The Weak Pigeonhole Principle

We prove that any regular resolution proof for the weak pigeon hole principle, with n holes and any number of pigeons, is of length , (for some global constant > 0).* Research supported by NSF grant CCR-9820831, US-Israel BSF grant 98-00349, and an NSERC grant. Research supported by US-Israel BSF grant 98-00349, and NSF grant CCR-9987077.     

Read-Once Branching Programs, Rectangular Proofs of the Pigeonhole Principle and the Transversal Calculus

We investigate read-once branching programs for the following search problem: given a Boolean m × n matrix with m > n, nd either an all-zero row, or two 1s in some column. Our primary motivation is that this models regular resolution proofs of the pigeonhole principle , and that for m > n ² no lower bounds are known for the length of such proofs. We prove exponential lower bounds (for arbitrarily large m!) if we further restrict this model by requiring the branching program either to finish one row of queries before asking queries about another row (the row model) or put the dual column restriction (the column model).Then we investigate a special class of resolution proofs for that operate with positive clauses of rectangular shape; we call this fragment the rectangular calculus. We show that all known upper bounds on the size of resolution proofs of actually give rise to proofs in this calculus and, inspired by this fact, also give a remarkably simple rectangular reformulation of the Haken–Buss–Turán lower bound for the case m n ². Finally we show that the rectangular calculus is equivalent to the column model on the one hand, and to transversal calculus on the other hand, where the latter is a natural proof system for estimating from below the transversal size of set families. In particular, our exponential lower bound for the column model translates both to the rectangular and transversal calculi.* Part of the work was done while this author was visiting Special Year on Logic and Algorithms at DIMACS, Princeton. Also supported by Russian Basic Research Foundation grant 96-01-01222. Part of this work was done while on sabbatical leave at the Institute for Advanced Study and Princeton University, Princeton. This work was supported by USA-Israel BSF grant 92-00106 and by a Wolfson research award administered by the Israeli Academy of Sciences, as well as a Sloan Foundation grant. This work was supported in part by National Science Foundation and DARPA under grant CCR-9627819, and by USA-Israel BSF grant 92-00106.     

Relativization makes contradictions harder for Resolution

We provide a number of simplified and improved separations between pairs of Resolution-with-bounded-conjunction refutation systems, Res(d)$\mathrm{Res}(d)$, as well as their tree-like versions, Res^?(d)${\mathrm{Res}}^{?}(d)$. The contradictions we use are natural combinatorial principles: the Least number principle  , _LNPn${\mathrm{LNP}}_n$ and an ordered variant thereof, the Induction principle  , _IPn${\mathrm{IP}}_n$.     

On The Parameterized Intractability Of Motif Search Problems*

We show that Closest Substring, one of the most important problems in the field of consensus string analysis, is W[1]-hard when parameterized by the number k of input strings (and remains so, even over a binary alphabet). This is done by giving a “strongly structure-preserving” reduction from the graph problem Clique to Closest Substring. This problem is therefore unlikely to be solvable in time O(f(k)•n^c) for any function f of k and constant c independent of k, i.e., the combinatorial explosion seemingly inherent to this NP-hard problem cannot be restricted to parameter k. The problem can therefore be expected to be intractable, in any practical sense, for k ≥ 3. Our result supports the intuition that Closest Substring is computationally much harder than the special case of Closest String, althoughb othp roblems are NP-complete. We also prove W[1]-hardness for other parameterizations in the case of unbounded alphabet size. Our W[1]-hardness result for Closest Substring generalizes to Consensus Patterns, a problem arising in computational biology. * An extended abstract of this paper was presented at the 19th International Symposium on Theoretical Aspects of Computer Science (STACS 2002), Springer-Verlag, LNCS 2285, pages 262–273, held in Juan-Les-Pins, France, March 14–16, 2002. † Work was supported by the Deutsche Forschungsgemeinschaft (DFG), research project “OPAL” (optimal solutions for hard problems in computational biology), NI 369/2. ‡ Work was done while the author was with Wilhelm-Schickard-Institut für Informatik, Universit?t Tübingen. Work was partially supported by the Deutsche Forschungsgemeinschaft (DFG), Emmy Noether research group “PIAF” (fixed-parameter algorithms), NI 369/4.     

A Computationally Motivated Definition Of Parametric Estimation And Its Applications To The Gaussian Distribution

We introduce a treatment of parametric estimation in which optimality of an estimator is measured in probability rather than in variance (the measure for which the strongest general results are known in statistics). Our motivation is that the quality of an approximation algorithm is measured by the probability that it fails to approximate the desired quantity within a set tolerance. We concentrate on the Gaussian distribution and show that the sample mean is the unique “best” estimator, in probability, for the mean of a Gaussian distribution. We also extend this method to general penalty functions and to multidimensional spherically symmetric Gaussians. The algorithmic significance of studying the Gaussian distribution is established by showing that determining the average matching size in a graph is #P-hard, and moreover approximating it reduces to estimating the mean of a random variable that (under some mild conditions) has a distribution closely approximating a Gaussian. This random variable is (essentially) polynomial time samplable, thereby yielding an FPRAS for the problem.     

On the computational complexity of cut-reduction

Using appropriate notation systems for proofs, cut-reduction can often be rendered feasible on these notations. Explicit bounds can be given. Developing a suitable notation system for Bounded Arithmetic, and applying these bounds, all the known results on definable functions of certain such theories can be reobtained in a uniform way.     

Short propositional refutations for dense random 3CNF formulas

Random 3CNF formulas constitute an important distribution for measuring the average-case behavior of propositional proof systems. Lower bounds for random 3CNF refutations in many propositional proof systems are known. Most notable are the exponential-size resolution refutation lower bounds for random 3CNF formulas with Ω(n^1.5−ε)$\mathrm{\Omega }(n^{1.5-\epsilon })$ clauses (Chvátal and Szemerédi [14], Ben-Sasson and Wigderson [10]). On the other hand, the only known non-trivial upper bound on the size of random 3CNF refutations in a non-abstract propositional proof system is for resolution with Ω(n²/log?n)$\mathrm{\Omega }(n^2/\log ?n)$ clauses, shown by Beame et al. [6]. In this paper we show that already standard propositional proof systems, within the hierarchy of Frege proofs, admit short refutations for random 3CNF formulas, for sufficiently large clause-to-variable ratio. Specifically, we demonstrate polynomial-size propositional refutations whose lines are TC⁰${\mathit{T}\mathit{C}}^0$ formulas (i.e., TC⁰${\mathit{T}\mathit{C}}^0$-Frege proofs) for random 3CNF formulas with n   variables and Ω(n^1.4)$\mathrm{\Omega }(n^{1.4})$ clauses.     

Combings of groups and the grammar of reparameterization

A new construction of combings is used to distinguish between several previously indistinguishable classes of groups associated to the theory of automatic groups and non-positive curvature in group theory. We construct synchronously bounded combings for a class of groups that are neither bicombable nor automatic. The linguistic complexity of these combings is analysed: in many cases the language of words in the combing is an indexed language.     

Logical omniscience as infeasibility

Logical theories for representing knowledge are often plagued by the so-called Logical Omniscience Problem. The problem stems from the clash between the desire to model rational agents, which should be capable of simple logical inferences, and the fact that any logical inference, however complex, almost inevitably consists of inference steps that are simple enough. This contradiction points to the fruitlessness of trying to solve the Logical Omniscience Problem qualitatively if the rationality of agents is to be maintained. We provide a quantitative solution to the problem compatible with the two important facets of the reasoning agent: rationality and resource boundedness. More precisely, we provide a test for the logical omniscience problem in a given formal theory of knowledge. The quantitative measures we use are inspired by the complexity theory. We illustrate our framework with a number of examples ranging from the traditional implicit representation of knowledge in modal logic to the language of justification logic, which is capable of spelling out the internal inference process. We use these examples to divide representations of knowledge into logically omniscient and not logically omniscient, thus trying to determine how much information about the reasoning process needs to be present in a theory to avoid logical omniscience.     

Monoid generalizations of the Richard Thompson groups

The groups G_k,1 of Richard Thompson and Graham Higman can be generalized in a natural way to monoids, that we call M_k,1, and to inverse monoids, called ; this is done by simply generalizing bijections to partial functions or partial injective functions. The monoids M_k,1 have connections with circuit complexity (studied in other papers). Here we prove that M_k,1 and are congruence-simple for all k. Their Green relations J and D are characterized: M_k,1 and are J-0-simple, and they have k−1 non-zero D-classes. They are submonoids of the multiplicative part of the Cuntz algebra O_k. They are finitely generated, and their word problem over any finite generating set is in P. Their word problem is coNP-complete over certain infinite generating sets.     

Characterizing Valiant's algebraic complexity classes

Valiant introduced 20 years ago an algebraic complexity theory to study the complexity of polynomial families. The basic computation model used is the arithmetic circuit, which makes these classes very easy to define and open to combinatorial techniques. In this paper we gather known results and new techniques under a unifying theme, namely the restrictions imposed upon the gates of the circuit, building a hierarchy from formulas to circuits. As a consequence we get simpler proofs for known results such as the equality of the classes VNP and _VNPe${\mathrm{VNP}}_{\mathrm{e}}$ or the completeness of the Determinant for VQP, and new results such as a characterization of the classes VQP and VP (which we can also apply to the Boolean class LOGCFL) or a full answer to a conjecture in Bürgisser's book [Completeness and reduction in algebraic complexity theory, Algorithms and Computation in Mathematics, vol. 7, Springer, Berlin, 2000]. We also show that for circuits of polynomial depth and unbounded size these models all have the same expressive power and can be used to characterize a uniform version of VNP.     

The axiomatic power of Kolmogorov complexity

The famous Gödel incompleteness theorem states that for every consistent, recursive, and sufficiently rich formal theory T there exist true statements that are unprovable in T. Such statements would be natural candidates for being added as axioms, but how can we obtain them? One classical (and well studied) approach is to add to some theory T an axiom that claims the consistency of T  . In this paper we discuss another approach motivated by Chaitin's version of Gödel's theorem where axioms claiming the randomness (or incompressibility) of some strings are probabilistically added, and show that it is not really useful, in the sense that this does not help us prove new interesting theorems. This result answers a question recently asked by Lipton. The situation changes if we take into account the size of the proofs: randomly chosen axioms may help making proofs much shorter (unless NP=PSPACE$\mathrm{NP}=\mathrm{PSPACE}$).     

A feasible theory of truth over combinatory algebra

We define an applicative theory of truth _TPT${\mathsf{T}}_{\mathsf{PT}}$ which proves totality exactly for the polynomial time computable functions. _TPT${\mathsf{T}}_{\mathsf{PT}}$ has natural and simple axioms since nearly all its truth axioms are standard for truth theories over an applicative framework. The only exception is the axiom dealing with the word predicate. The truth predicate can only reflect elementhood in the words for terms that have smaller length than a given word. This makes it possible to achieve the very low proof-theoretic strength. Truth induction can be allowed without any constraints. For these reasons the system _TPT${\mathsf{T}}_{\mathsf{PT}}$ has the high expressive power one expects from truth theories. It allows embeddings of feasible systems of explicit mathematics and bounded arithmetic.     

Cut elimination for a calculus with context-dependent rules

Context-dependent rules are an obstacle to cut elimination. Turning to a generalised sequent style formulation using deep inferences is helpful, and for the calculus presented here it is essential. Cut elimination is shown for a substructural, multiplicative, pure propositional calculus. Moreover we consider the extra problems induced by non-logical axioms and extend the results to additive connectives and quantifiers. Received: 11 April 1998 / Published online: 25 January 2001     

Cut-free formulations for a quantified logic of here and there

A predicate extension SQHT⁼ of the logic of here-and-there was introduced by V. Lifschitz, D. Pearce, and A. Valverde to characterize strong equivalence of logic programs with variables and equality with respect to stable models. The semantics for this logic is determined by intuitionistic Kripke models with two worlds (here and there) with constant individual domain and decidable equality. Our sequent formulation has special rules for implication and for pushing negation inside formulas. The soundness proof allows us to establish that SQHT⁼ is a conservative extension of the logic of weak excluded middle with respect to sequents without positive occurrences of implication. The completeness proof uses a non-closed branch of a proof search tree. The interplay between rules for pushing negation inside and truth in the “there” (non-root) world of the resulting Kripke model can be of independent interest. We prove that existence is definable in terms of remaining connectives.     

On The Approximability Of The Traveling Salesman Problem

We show that the traveling salesman problem with triangle inequality cannot be approximated with a ratio better than when the edge lengths are allowed to be asymmetric and when the edge lengths are symmetric, unless P=NP. The best previous lower bounds were and respectively. The reduction is from H?stad’s maximum satisfiability of linear equations modulo 2, and is nonconstructive. * Supported in part by NSF ITR Grant CCR-0121555. † Supported by NSF award CCR-0307536 and a Sloan foundation fellowship.     

A restricted computation model on Scott domains and its partial primitive recursive functionals

The paper builds on both a simply typed term system and a computation model on Scott domains via so-called parallel typed while programs (PTWP). The former provides a notion of partial primitive recursive functional on Scott domains supporting a suitable concept of parallelism. Computability on Scott domains seems to entail that Kleene's schema of higher type simultaneous course-of-values recursion (scvr) is not reducible to partial primitive recursion. So extensions and PTWP are studied that are closed under scvr. The twist are certain type 1 G?del recursors for simultaneous partial primitive recursion. Formally, denotes a function , however, is modelled such that is finite, or in other words, a partial sequence. As for PTWP, the concept of type writable variables is introduced, providing the possibility of creating and manipulating partial sequences. It is shown that the PTWP-computable functionals coincide with those definable in plus a constant for sequential minimisation. In particular, the functionals definable in denoted can be characterised by a subclass of PTWP-computable functionals denoted . Moreover, hierarchies of strictly increasing classes in the style of Heinermann and complexity classes are introduced such that . These results extend those for and PTWP [Nig94]. Finally, scvr is employed to define for each type the enumeration functional of all finite elements of . Received January 30, 1996     

Subrecursive functions on partial sequences

The paper studies a domain theoretical notion of primitive recursion over partial sequences in the context of Scott domains. Based on a non-monotone coding of partial sequences, this notion supports a rich concept of parallelism in the sense of Plotkin. The complexity of these functions is analysed by a hierarchy of classes similar to the Grzegorczyk classes. The functions considered are characterised by a function algebra generated by continuity preserving operations starting from computable initial functions. Its layers are related to those above by showing , thus generalising results of Schwichtenberg/Müller and Niggl. Received: 18 November 1996