The model of span programs is a linear algebraic model of
computation. Lower bounds for span programs imply lower bounds for
contact schemes, symmetric branching programs and for formula size.
Monotone span programs correspond also to linear secret-sharing schemes.
We present a new technique for proving lower bounds for ... more >>>

In unrestricted branching programs all variables may be tested
arbitrarily often on each path. But exponential lower bounds are only
known, if on each path the number of tests of each variable is bounded
(Borodin, Razborov and Smolensky (1993)). We examine branching programs
in which for each path the ... more >>>

The Steiner tree problem requires to find a shortest tree connection
a given set of terminal points in a metric space. We suggest a better
and fast heuristic for the Steiner problem in graphs and in
rectilinear plane. This heuristic finds a Steiner tree at ... more >>>

It was shown some years ago that the computation time for many important
Boolean functions of n arguments on concurrent-read exclusive-write
parallel random-access machines
(CREW PRAMs) of unlimited size is at least f(n) = 0.72 log n.
On the other hand, it ... more >>>

This paper tries to fully characterize the properties and relationships
of space classes defined by Turing machines that use less than
logarithmic space - may they be deterministic,
nondeterministic or alternating (DTM, NTM or ATM).

We provide several examples of specific languages ... more >>>

This paper proves that if strong pseudorandom number generators or
one-way functions exist, then the class of languages that have
polynomial-sized circuits is not small within exponential
time, in terms of the resource-bounded measure theory of Lutz.
More precisely, if for some \epsilon > 0 there exist nonuniformly
2^{n^{\epsilon}}-hard ... more >>>

The {\it nucleon} is introduced as a new allocation concept for
non-negative cooperative n-person transferable utility games.
The nucleon may be viewed as the multiplicative analogue of
Schmeidler's nucleolus.
It is shown that the nucleon of (not necessarily bipartite) matching
games ... more >>>

It is shown that decomposition via Chinise Remainder does not
yield polynomial size depth 3 threshold circuits for iterated
multiplication of n-bit numbers. This result is achieved by
proving that, in contrast to multiplication of two n-bit
numbers, powering, division, and other related problems, the
... more >>>

We prove an unexpected upper bound on a communication game proposed
by Jeff Edmonds and Russell Impagliazzo as an approach for
proving lower bounds for time-space tradeoffs for branching programs.
Our result is based on a generalization of a construction of Erdos,
Frankl and Rodl of a large 3-hypergraph ... more >>>

The graph homomorphism problem is a canonical $NP$-complete problem.
It generalizes various other well-studied problems such as graph
coloring and finding cliques. To get a better insight into a
combinatorial problem, one often studies relaxations of the problem.
We define fractional homomorphisms and pseudo-homomorphisms ... more >>>

The parallel repetition conjecture (PRC) of Feige and Lovasz says that the
error probability of a two prover one round interactive protocol repeated $n$
times in parallel is exponentially small in $n$.
We show that the PRC is true in the case when
the bipartite graph of dependence between ... more >>>

We show that, for fixed dimension $n$, the approximation of
inner and outer $j$-radii of polytopes in ${\Re}^n$, endowed
with the Euclidean norm, is polynomial.

We show what happen to learning if the learner can use NP-oracle.
A consequence of our results we show that
If NP\subset P/poly then the polynomial Hierarchy collapses to ZPP^NP

END_OF_DESCRIPTION
Contact: bshouty@cpsc.ucalgary.ca (Nader Bshouty)

We consider the problem of fair cost allocation for traveling
salesman games for which the triangle inequality holds. We
give examples showing that the core of such games may be
empty, even for the case of Euclidean distances. On the
positive side, we develop an LP-based ... more >>>

We consider the threshold circuit complexity of computing
the multiple product modulo m in threshold circuits.

We show that polynomially rankable distributions
do not provide harder instances than uniform distributions
for NP problems. In particular, we show that if Levin's
randomized tiling problem is solvable in polynomial time on
average, then every NP problem under any p-rankable
... more >>>

We consider a dynamic fault diagnosis problem: There are n
processors, to be tested in a series of rounds. In every
testing round we use a directed matching to have some
processors report on the status (good or faulty) of other
processors. Also in each round ... more >>>

In contrast to deterministic or nondeterministic computation, it is
a fundamental open problem in randomized computation how to separate
different randomized time classes (at this point we do not even know
how to separate linear randomized time from ${\mathcal O}(n^{\log n})$
randomized time) or how to ... more >>>

We consider strings which are succinctly described. The description
is in terms of straight-line programs in which the constants are
symbols and the only operation is the concatenation. Such
descriptions correspond to the systems of recurrences or to
context-free grammars generating single words. The descriptive ... more >>>

We attempt to reconcile the two distinct views of approximation
classes: syntactic and computational.
Syntactic classes such as MAX SNP allow for clean complexity-theoretic
results and natural complete problems, while computational classes such
as APX allow us to work with problems whose approximability is
well-understood. Our results give a computational ... more >>>

This paper continues the investigation of the connection between proof
systems and approximation. The emphasis is on proving ``tight''
non-approximability results via consideration of measures like the
``free bit complexity'' and the ``amortized free bit complexity'' of
proof systems.

The first part of the paper presents a collection of new ... more >>>

In this paper the $R$-machines defined by Blum, Shub and Smale
are generalized by allowing infinite convergent computations.
The description of real numbers is infinite.
Therefore, considering arithmetic operations on real numbers should
also imply infinite computations on {\em analytic machines}.
We prove that $\R$-computable functions are $\Q$-analytic.
We show ... more >>>

We introduce new algorithms for lattice basis reduction that are
improvements of the LLL-algorithm. We demonstrate the power of
these algorithms by solving random subset sum problems of
arbitrary density with 74 and 82 many weights, by breaking the
Chor-Rivest cryptoscheme in dimensions 103 and 151 ... more >>>

We say an integer polynomial $p$, on Boolean inputs, weakly
$m$-represents a Boolean function $f$ if $p$ is non-constant and is zero (mod
$m$), whenever $f$ is zero. In this paper we prove that if a polynomial
weakly $m$-represents the Mod$_q$ function on $n$ inputs, where $q$ and $m$
more >>>

In (Allender and Strauss, FOCS '95), we defined a notion of
measure on the complexity class P (in the spirit of the work of (Lutz,
JCSS '92) that provides a notion of measure on complexity classes at least
as large as E, and the work of (Mayordomo, Phd. ... more >>>

We show how to construct length-preserving 1-1 one-way
functions based on popular intractability assumptions (e.g., RSA, DLP).
Such 1-1 functions should not
be confused with (infinite) families of (finite) one-way permutations.
What we want and obtain is a single (infinite) 1-1 one-way function.

The Steiner tree problem asks for the shortest tree connecting
a given set of terminal points in a metric space. We design
new approximation algorithms for the Steiner tree problems
using a novel technique of choosing Steiner points in dependence
on the possible deviation from ... more >>>

Improving a long standing result of Sch\"{o}nhage, Paterson
and Pippenger we show that the MEDIAN of a set containing n
elements can always be found using at most 2.95n comparisons.

This is the full version of the paper. An extended abstract
version ... more >>>

We consider worst case time bounds for NP-complete problems
including 3-SAT, 3-coloring, 3-edge-coloring, and 3-list-coloring.
Our algorithms are based on a common generalization of these problems,
called symbol-system satisfiability or, briefly, SSS [R. Floyd &
R. Beigel, The Language of Machines]. 3-SAT is equivalent to
(2,3)-SSS while the other problems ... more >>>

We investigate the probabilistic communication complexity
(more exactly, the majority communication complexity,)
of the graph accessibility problem GAP and
its counting versions MOD_k-GAP, k > 1.
Due to arguments concerning matrix variation ranks
and certain projection reductions, we prove
that, for any partition of the input variables,
more >>>

We classify the univariate functions that are relativizable
closure properties of GapP, solving a problem posed by Hertrampf,
Vollmer, and Wagner (Structures '95). We also give a simple proof of
their classification of univariate functions that are relativizable
closure properties of #P.

For a set A and a number n let F_n^A(x_1,...,x_n) =
A(x_1)\cdots A(x_n). We study how hard it is to approximate this
function in terms of the number of queries required. For a general
set A we have exact bounds that depend on functions from coding
theory. These are applied ... more >>>

Identify a string x over {0,1} with the positive integer
whose binary representation is 1x. We say that a self-reduction is
k-local if on input x all queries belong to {x-1,...,x-k}. We show
that all k-locally self-reducible sets belong to PSPACE. However, the
power of k-local self-reductions changes drastically between ... more >>>

We consider noninteractive zero-knowledge proofs in the shared random
string model proposed by Blum, Feldman and Micali \cite{bfm}. Until
recently there was a sizable polynomial gap between the most
efficient noninteractive proofs for {\sf NP} based on general
complexity assumptions \cite{fls} versus those based on specific
algebraic assumptions \cite{Da}. ... more >>>

This paper studies distributions which
can be ``approximated'' by sampling algorithms in time polynomial in
the length of their outputs. First, it is known that if
polynomial-time samplable distributions are polynomial-time
computable, then NP collapses to P. This paper shows by a simple
... more >>>

We study the complexity of finding the values and optimal strategies of
MEAN PAYOFF GAMES on graphs, a family of perfect information games
introduced by Ehrenfeucht and Mycielski and considered by Gurvich,
Karzanov and Khachiyan. We describe a pseudo-polynomial time algorithm
for the solution of such games, the decision ... more >>>

We prove an optimal bound on the Shannon function
$L(n,m,\epsilon)$ which describes the trade-off between the
circuit-size complexity and the degree of approximation; that is
$$L(n,m,\epsilon)\ =\
\Theta\left(\frac{m\epsilon^2}{\log(2 + m\epsilon^2)}\right)+O(n).$$
Our bound applies to any partial boolean function
and any ... more >>>

Computational complexity is concerned with the complexity of solving
problems and computing functions and not with the complexity of verifying
circuit designs.
The importance of formal circuit verification is evident.
Therefore, a framework of a complexity theory for formal circuit
verification with binary decision diagrams ... more >>>

We investigate the phenomenon of depth-reduction in commutative
and non-commutative arithmetic circuits. We prove that in the
commutative setting, uniform semi-unbounded arithmetic circuits of
logarithmic depth are as powerful as uniform arithmetic circuits of
polynomial degree; earlier proofs did not work in the ... more >>>

We introduce the model of conservative one-way multiparty complexity
and prove lower and upper bounds on the complexity of pointer jumping.

The pointer jumping function takes as its input a directed layered
graph with a starting node and layers of nodes, and a single edge ... more >>>

A pseudo-random function is a fundamental cryptographic primitive
that is essential for encryption, identification and authentication.
We present a new cryptographic primitive called pseudo-random
synthesizer and show how to use it in order to get a
parallel construction of a pseudo-random function.
We show an $NC^1$ implementation of synthesizers ... more >>>

We examine the power of Boolean functions with low L_1 norms in several
settings. In large part of the recent literature, the degree of a polynomial
which represents a Boolean function in some way was chosen to be the measure of the complexity of the Boolean function.
However, some functions ... more >>>

An increasing number of results in graph theory, combinatorics and
theoretical computer science is obtained with the help of computers,
e.g. the proof of the Four Colours Theorem or the computation of
certain Ramsey numbers. Binary decision diagrams, known as tools in
hardware verification ... more >>>

In this paper, we present stronger results in the theory of succinct
problem representation by boolean circuits, and establish a close
relationship between succinct problems and leaf languages. As
a major tool, we use quantifierfree projection reductions
from descriptive complexity theory.

In particular, we show that the succinct upgrading ... more >>>

This paper provides logspace and small circuit depth analogs
of the result of Valiant-Vazirani, which is a randomized (or
nonuniform) reduction from NP to its arithmetic analog ParityP.
We show a similar randomized reduction between the
Boolean classes NL and semi-unbounded fan-in Boolean circuits and
their arithmetic counterparts. These ... more >>>

The main result of this paper is a Omega(n^{1/4}) lower bound
on the size of a sigmoidal circuit computing a specific AC^0_2 function.
This is the first lower bound for the computation model of sigmoidal
circuits with unbounded weights. We also give upper and lower bounds for
the ... more >>>

In this primarily expository
paper, we discuss the connections between two popular and useful
tools in theoretical computer science, namely,
universal hashing and pairwise
independent random variables; and classical combinatorial stuctures
such as error-correcting codes, balanced incomplete block designs,
difference matrices
... more >>>

We establish significantly improved bounds on the performance of the greedy
algorithm for approximating MINIMUM SET COVER and MINIMUM PARTIAL COVER. Our
improvements result from a new approach to both problems. In particular,
(a) we improve the known bound on the performance ratio of the greedy ... more >>>

We define the notion of a randomized branching program in
the natural way similar to the definition of a randomized
circuit. We exhibit an explicit function $f_{n}$ for which
we prove that:
1) $f_{n}$ can be computed by polynomial size randomized
... more >>>

We introduce a new method for proving explicit upper bounds
on the VC Dimension of general functional basis networks,
and prove as an application, for the first time, that the
VC Dimension of analog neural networks with the sigmoidal
activation function $\sigma(y)=1/1+e^{-y}$ ... more >>>

We provide an exposition of three Lemmas which relate
general properties of distributions
with the exclusive-or of certain bit locations.

The first XOR-Lemma, commonly attributed to U.V. Vazirani,
relates the statistical distance of a distribution from uniform
to the maximum bias of the xor of certain bit positions.
more >>>

We prove an exponential lower bound on the size of any
fixed-degree algebraic decision tree for solving MAX, the
problem of finding the maximum of $n$ real numbers. This
complements the $n-1$ lower bound of Rabin \cite{R72} on
the depth of ... more >>>

We deal with the problem of extracting as much randomness as possible
from a defective random source.
We devise a new tool, a ``merger'', which is a function that accepts
d strings, one of which is uniformly distributed,
and outputs a single string that is guaranteed ... more >>>

We show that a DNF formula that has a CNF representation that contains
at least one ``$1/poly$-heavy''
clause with respect to a distribution $D$ is weakly learnable
under this distribution. So DNF that are not weakly
learnable under the distribution $D$ do not have
any ``$1/poly$-heavy'' clauses in any of ... more >>>

We present a $2^{\tilde O(\sqrt{n})}$ time exact learning
algorithm for polynomial size
DNF using equivalence queries only. In particular, DNF
is PAC-learnable in subexponential time under any distribution.
This is the first subexponential time
PAC-learning algorithm for DNF under any distribution.

We show that hitting sets can derandomize any BPP-algorithm.
This gives a positive answer to a fundamental open question in
probabilistic algorithms. More precisely, we present a polynomial
time deterministic algorithm which uses any given hitting set
to approximate the fractions of 1's in the ... more >>>

Consider a problem involving updates and queries of a data structure.
Assume that there exists a family of algorithms which exhibit a
tradeoff between query and update time. We demonstrate a general
technique of constructing from such a family
a single algorithm with best amortized time. We indicate some ... more >>>

We extend the lower bounds on the depth of algebraic decision trees
to the case of {\em randomized} algebraic decision trees (with
two-sided error) for languages being finite unions of hyperplanes
and the intersections of halfspaces, solving a long standing open
problem. As an application, among ... more >>>