arXiv subjects

Valmir C. Barbosa

Publications and source records attributed to Valmir C. Barbosa.

At least 55 records · Page 3Linked to original sources

arXiv2011

Evolved preambles for MAX-SAT heuristics

MAX-SAT heuristics normally operate from random initial truth assignments to the variables. We consider the use of what we call preambles, which are sequences of variables with corresponding single-variable assignment actions intended to be used to determine a more suitable initial truth assignment for a given problem instance and a given heuristic. For a number of well established MAX-SAT heuristics and benchmark instances, we demonstrate that preambles can be evolved by a genetic algorithm such that the heuristics are outperformed in a significant fraction of the cases.

cs.AI↗

arXiv2010

Network algorithmics and the emergence of information integration in cortical models

An information-theoretic framework known as integrated information theory (IIT) has been introduced recently for the study of the emergence of consciousness in the brain [D. Balduzzi and G. Tononi, PLoS Comput. Biol. 4, e1000091 (2008)]. IIT purports that this phenomenon is to be equated with the generation of information by the brain surpassing the information which the brain's constituents already generate independently of one another. IIT is not fully plausible in its modeling assumptions, nor is it testable due to severe combinatorial growth embedded in its key definitions. Here we introduce an alternative to IIT which, while inspired in similar information-theoretic principles, seeks to address some of IIT's shortcomings to some extent. Our alternative framework uses the same network-algorithmic cortical model we introduced earlier [A. Nathan and V. C. Barbosa, Phys. Rev. E 81, 021916 (2010)] and, to allow for somewhat improved testability relative to IIT, adopts the well-known notions of information gain and total correlation applied to a set of variables representing the reachability of neurons by messages in the model's dynamics. We argue that these two quantities relate to each other in such a way that can be used to quantify the system's efficiency in generating information beyond that which does not depend on integration, and give computational results on our cortical model and on variants thereof that are either structurally random in the sense of an Erdos-Renyi random directed graph or structurally deterministic. We have found that our cortical model stands out with respect to the others in the sense that many of its instances are capable of integrating information more efficiently than most of those others' instances.

q-bio.NC↗

arXiv2010

Revisiting deadlock prevention: a probabilistic approach

We revisit the deadlock-prevention problem by focusing on priority digraphs instead of the traditional wait-for digraphs. This has allowed us to formulate deadlock prevention in terms of prohibiting the occurrence of directed cycles even in the most general of wait models (the so-called AND-OR model, in which prohibiting wait-for directed cycles is generally overly restrictive). For a particular case in which the priority digraphs are somewhat simplified, we introduce a Las Vegas probabilistic mechanism for resource granting and analyze its key aspects in detail.

cs.DC↗

arXiv2010

Early appraisal of the fixation probability in directed networks

In evolutionary dynamics, the probability that a mutation spreads through the whole population, having arisen in a single individual, is known as the fixation probability. In general, it is not possible to find the fixation probability analytically given the mutant's fitness and the topological constraints that govern the spread of the mutation, so one resorts to simulations instead. Depending on the topology in use, a great number of evolutionary steps may be needed in each of the simulation events, particularly in those that end with the population containing mutants only. We introduce two techniques to accelerate the determination of the fixation probability. The first one skips all evolutionary steps in which the number of mutants does not change and thereby reduces the number of steps per simulation event considerably. This technique is computationally advantageous for some of the so-called layered networks. The second technique, which is not restricted to layered networks, consists of aborting any simulation event in which the number of mutants has grown beyond a certain threshold value, and counting that event as having led to a total spread of the mutation. For large populations, and regardless of the network's topology, we demonstrate, both analytically and by means of simulations, that using a threshold of about 100 mutants leads to an estimate of the fixation probability that deviates in no significant way from that obtained from the full-fledged simulations. We have observed speedups of two orders of magnitude for layered networks with 10000 nodes.

q-bio.PE↗

arXiv2010

Network conduciveness with application to the graph-coloring and independent-set optimization transitions

We introduce the notion of a network's conduciveness, a probabilistically interpretable measure of how the network's structure allows it to be conducive to roaming agents, in certain conditions, from one portion of the network to another. We exemplify its use through an application to the two problems in combinatorial optimization that, given an undirected graph, ask that its so-called chromatic and independence numbers be found. Though NP-hard, when solved on sequences of expanding random graphs there appear marked transitions at which optimal solutions can be obtained substantially more easily than right before them. We demonstrate that these phenomena can be understood by resorting to the network that represents the solution space of the problems for each graph and examining its conduciveness between the non-optimal solutions and the optimal ones. At the said transitions, this network becomes strikingly more conducive in the direction of the optimal solutions than it was just before them, while at the same time becoming less conducive in the opposite direction. We believe that, besides becoming useful also in other areas in which network theory has a role to play, network conduciveness may become instrumental in helping clarify further issues related to NP-hardness that remain poorly understood.

cond-mat.stat-mech↗

arXiv2009

Network algorithmics and the emergence of the cortical synaptic-weight distribution

When a neuron fires and the resulting action potential travels down its axon toward other neurons' dendrites, the effect on each of those neurons is mediated by the weight of the synapse that separates it from the firing neuron. This weight, in turn, is affected by the postsynaptic neuron's response through a mechanism that is thought to underlie important processes such as learning and memory. Although of difficult quantification, cortical synaptic weights have been found to obey a long-tailed unimodal distribution peaking near the lowest values, thus confirming some of the predictive models built previously. These models are all causally local, in the sense that they refer to the situation in which a number of neurons all fire directly at the same postsynaptic neuron. Consequently, they necessarily embody assumptions regarding the generation of action potentials by the presynaptic neurons that have little biological interpretability. In this letter we introduce a network model of large groups of interconnected neurons and demonstrate, making none of the assumptions that characterize the causally local models, that its long-term behavior gives rise to a distribution of synaptic weights with the same properties that were experimentally observed. In our model the action potentials that create a neuron's input are, ultimately, the product of network-wide causal chains relating what happens at a neuron to the firings of others. Our model is then of a causally global nature and predicates the emergence of the synaptic-weight distribution on network structure and function. As such, it has the potential to become instrumental also in the study of other emergent cortical phenomena.

q-bio.NC↗

arXiv2009

6-cycle double covers of cubic graphs

A cycle double cover (CDC) of an undirected graph is a collection of the graph's cycles such that every edge of the graph belongs to exactly two cycles. We describe a constructive method for generating all the cubic graphs that have a 6-CDC (a CDC in which every cycle has length 6). As an application of the method, we prove that all such graphs have a Hamiltonian cycle. A sense of direction is an edge labeling on graphs that follows a globally consistent scheme and is known to considerably reduce the complexity of several distributed problems. In [9], a particular instance of sense of direction, called a chordal sense of direction (CSD), is studied and the class of k-regular graphs that admit a CSD with exactly k labels (a minimal CSD) is analyzed. We now show that nearly all the cubic graphs in this class have a 6-CDC, the only exception being K4.

cs.DM↗

arXiv2008

Approximate conditional distributions of distances between nodes in a two-dimensional sensor network

When we represent a network of sensors in Euclidean space by a graph, there are two distances between any two nodes that we may consider. One of them is the Euclidean distance. The other is the distance between the two nodes in the graph, defined to be the number of edges on a shortest path between them. In this paper, we consider a network of sensors placed uniformly at random in a two-dimensional region and study two conditional distributions related to these distances. The first is the probability distribution of distances in the graph, conditioned on Euclidean distances; the other is the probability density function associated with Euclidean distances, conditioned on distances in the graph. We study these distributions both analytically (when feasible) and by means of simulations. To the best of our knowledge, our results constitute the first of their kind and open up the possibility of discovering improved solutions to certain sensor-network problems, as for example sensor localization.

cs.NI↗

arXiv2008

Network growth for enhanced natural selection

Natural selection and random drift are competing phenomena for explaining the evolution of populations. Combining a highly fit mutant with a population structure that improves the odds that the mutant spreads through the whole population tips the balance in favor of natural selection. The probability that the spread occurs, known as the fixation probability, depends heavily on how the population is structured. Certain topologies, albeit highly artificially contrived, have been shown to exist that favor fixation. We introduce a randomized mechanism for network growth that is loosely inspired in some of these topologies' key properties and demonstrate, through simulations, that it is capable of giving rise to structured populations for which the fixation probability significantly surpasses that of an unstructured population. This discovery provides important support to the notion that natural selection can be enhanced over random drift in naturally occurring population structures.

q-bio.PE↗

arXiv2007

Descents and nodal load in scale-free networks

The load of a node in a network is the total traffic going through it when every node pair sustains a uniform bidirectional traffic between them on shortest paths. We show that nodal load can be expressed in terms of the more elementary notion of a node's descents in breadth-first-search (BFS or shortest-path) trees, and study both the descent and nodal-load distributions in the case of scale-free networks. Our treatment is both semi-analytical (combining a generating-function formalism with simulation-derived BFS branching probabilities) and computational for the descent distribution; it is exclusively computational in the case of the load distribution. Our main result is that the load distribution, even though it can be disguised as a power-law through subtle (but inappropriate) binning of the raw data, is in fact a succession of sharply delineated probability peaks, each of which can be clearly interpreted as a function of the underlying BFS descents. This find is in stark contrast with previously held belief, based on which a power law of exponent -2.2 was conjectured to be valid regardless of the exponent of the power-law distribution of node degrees.

cond-mat.stat-mech↗

arXiv2007

Optimization of supply diversity for the self-assembly of simple objects in two and three dimensions

The field of algorithmic self-assembly is concerned with the design and analysis of self-assembly systems from a computational perspective, that is, from the perspective of mathematical problems whose study may give insight into the natural processes through which elementary objects self-assemble into more complex ones. One of the main problems of algorithmic self-assembly is the minimum tile set problem (MTSP), which asks for a collection of types of elementary objects (called tiles) to be found for the self-assembly of an object having a pre-established shape. Such a collection is to be as concise as possible, thus minimizing supply diversity, while satisfying a set of stringent constraints having to do with the termination and other properties of the self-assembly process from its tile types. We present a study of what we think is the first practical approach to MTSP. Our study starts with the introduction of an evolutionary heuristic to tackle MTSP and includes results from extensive experimentation with the heuristic on the self-assembly of simple objects in two and three dimensions. The heuristic we introduce combines classic elements from the field of evolutionary computation with a problem-specific variant of Pareto dominance into a multi-objective approach to MTSP.

cs.NE↗

arXiv2007

Emergence of scale-free behavior in networks from limited-horizon linking and cost trade-offs

We study network growth from a fixed set of initially isolated nodes placed at random on the surface of a sphere. The growth mechanism we use adds edges to the network depending on strictly local gain and cost criteria. Only nodes that are not too far apart on the sphere may be considered for being joined by an edge. Given two such nodes, the joining occurs only if the gain of doing it surpasses the cost. Our model is based on a multiplicative parameter lambda that regulates, in a function of node degrees, the maximum geodesic distance that is allowed between nodes for them to be considered for joining. For n nodes distributed uniformly on the sphere, and for lambda*sqrt(n) within limits that depend on cost-related parameters, we have found that our growth mechanism gives rise to power-law distributions of node degree that are invariant for constant lambda*sqrt(n). We also study connectivity- and distance-related properties of the networks.

cond-mat.stat-mech↗

arXiv2007

Reachability and recoverability of sink nodes in growing acyclic directed networks

We study the growth of networks from a set of isolated ground nodes by the addition of one new node per time step and also of a fixed number of directed edges leading from the new node to randomly selected nodes already in the network. A fixed-width time window is used so that, in general, only nodes that entered the network within the latest window may receive new incoming edges. The resulting directed network is acyclic at all times and allows some of the ground nodes, then called sinks, to be reached from some of the non-ground nodes. We regard such networks as representative of abstract systems of partially ordered constituents, for example in some of the domains related to technological evolution. Two properties of interest are the number of sinks that can be reached from a randomly chosen non-ground node (its reach) and, for a fixed sink, the number of nonoverlapping directed paths through which the sink can be reached, at a given time, from some of the latest nodes to have entered the network. We demonstrate, by means of simulations and also of analytic characterizations, that reaches are distributed according to a power law and that the desired directed paths are expected to occur in very small numbers, perhaps indicating that recovering sinks late in the process of network growth is strongly sensitive to accidental path disruptions.

cond-mat.stat-mech↗

arXiv2007

An algorithm for clock synchronization with the gradient property in sensor networks

We introduce a distributed algorithm for clock synchronization in sensor networks. Our algorithm assumes that nodes in the network only know their immediate neighborhoods and an upper bound on the network's diameter. Clock-synchronization messages are only sent as part of the communication, assumed reasonably frequent, that already takes place among nodes. The algorithm has the gradient property of [2], achieving an O(1) worst-case skew between the logical clocks of neighbors. As in the case of [3,8], the algorithm's actions are such that no constant lower bound exists on the rate at which logical clocks progress in time, and for this reason the lower bound of [2,5] that forbids constant skew between neighbors does not apply.

cs.DC↗

arXiv2007

V-like formations in flocks of artificial birds

We consider flocks of artificial birds and study the emergence of V-like formations during flight. We introduce a small set of fully distributed positioning rules to guide the birds' movements and demonstrate, by means of simulations, that they tend to lead to stabilization into several of the well-known V-like formations that have been observed in nature. We also provide quantitative indicators that we believe are closely related to achieving V-like formations, and study their behavior over a large set of independent simulations.

cs.NE↗

arXiv2007

On simulating nondeterministic stochastic activity networks

In this work we deal with a mechanism for process simulation called a NonDeterministic Stochastic Activity Network (NDSAN). An NDSAN consists basically of a set of activities along with precedence relations involving these activities, which determine their order of execution. Activity durations are stochastic, given by continuous, nonnegative random variables. The nondeterministic behavior of an NDSAN is based on two additional possibilities: (i) by associating choice probabilities with groups of activities, some branches of execution may not be taken; (ii) by allowing iterated executions of groups of activities according to predetermined probabilities, the number of times an activity must be executed is not determined a priori. These properties lead to a rich variety of activity networks, capable of modeling many real situations in process engineering, project design, and troubleshooting. We describe a recursive simulation algorithm for NDSANs, whose repeated execution produces a close approximation to the probability distribution of the completion time of the entire network. We also report on real-world case studies.

cs.DM↗

arXiv2006

Partially ordered distributed computations on asynchronous point-to-point networks

Asynchronous executions of a distributed algorithm differ from each other due to the nondeterminism in the order in which the messages exchanged are handled. In many situations of interest, the asynchronous executions induced by restricting nondeterminism are more efficient, in an application-specific sense, than the others. In this work, we define partially ordered executions of a distributed algorithm as the executions satisfying some restricted orders of their actions in two different frameworks, those of the so-called event- and pulse-driven computations. The aim of these restrictions is to characterize asynchronous executions that are likely to be more efficient for some important classes of applications. Also, an asynchronous algorithm that ensures the occurrence of partially ordered executions is given for each case. Two of the applications that we believe may benefit from the restricted nondeterminism are backtrack search, in the event-driven case, and iterative algorithms for systems of linear equations, in the pulse-driven case.

cs.DC↗

arXiv2005

Emergence of scale-free networks from local connectivity and communication trade-offs

We introduce a new mechanism of connectivity evolution in networks to account for the emergence of scale-free behavior. The mechanism works on a fixed set of nodes and promotes growth from a minimally connected initial topology by the addition of edges. A new edge is added between two nodes depending on the trade-off between a gain and a cost function of local connectivity and communication properties. We report on simulation results that indicate the appearance of power-law distributions of node degrees for selected parameter combinations.

cond-mat.stat-mech↗