Fabio Daolio

Evaluation of parallel particle swarm optimization algorithms within the CUDA TM architecture

Particle swarm optimization (PSO), like other population-based meta-heuristics, is intrinsically parallel and can be effectively implemented on Graphics Processing Units (GPUs), which are, in fact, massively parallel processing architectures. In this paper we discuss possible approaches to parallelizing PSO on graphics hardware within the Compute Unified Device Architecture (CUDA(TM)), a GPU programming environment by nVIDIA(TM) which supports the companys latest cards. In particular, two different ways of exploiting GPU parallelism are explored and evaluated. The execution speed of the two parallel algorithms is compared, on functions which are typically used as benchmarks for PSO, with a standard sequential implementation of PSO (SPSO), as well as with recently published results of other parallel implementations. An in-depth study of the computation efficiency of our parallel algorithms is carried out by assessing speed-up and scale-up with respect to SPSO. Also reported are some results about the optimization effectiveness of the parallel implementations with respect to SPSO, in cases when the parallel versions introduce some possibly significant difference with respect to the sequential version.

Smart rewiring for network robustness

While new forms of attacks are developed every day to compromise essential infrastructures, service providers are also expected to develop strategies to mitigate the risk of extreme failures. In this context, tools of network science have been used to evaluate network robustness and propose resilient topologies against attacks. We present here a new rewiring method to modify the network topology improving its robustness, based on the evolution of the network largest component during a sequence of targeted attacks. In comparison to previous strategies, our method lowers by several orders of magnitude the computational effort necessary to improve robustness. Our rewiring also drives the formation of layers of nodes with similar degree while keeping a highly modular structure. This modular onion-like structure is a particular class of the onion-like structure previously described in the literature. We apply our rewiring strategy to an unweighted representation of the World Air-transportation network and show that an improvement of thirty percent in its overall robustness can be achieved through smart swaps of around nine percent of its links.

GPU-Based Road Sign Detection Using Particle Swarm Optimization

Road Sign Detection is a major goal of Advanced Driving Assistance Systems (ADAS). Since the dawn of this discipline, much work based on different techniques has been published which shows that traffic signs can be first detected and then classified in video sequences in real time. While detection is usually performed using classical computer vision techniques based on color and/or shape matching, most often classification is performed by neural networks. In this work we present a novel approach based on both sign shape and color which uses Particle Swarm Optimization (PSO) for detection. Remarkably, a single fitness function can be used both to detect a sign belonging to a certain category and, at the same time, to estimate its actual position with respect to the camera reference frame. To speed up execution times, the algorithm exploits the parallelism offered by modern graphics cards and, in particular, the CUDA™ architecture by nVIDIA. The effectiveness of the approach has been assessed on a synthetic video sequence, which has been successfully processed in real time at full frame rate.

Local Optima Networks of the Quadratic Assignment Problem

Using a recently proposed model for combinatorial landscapes, Local Optima Networks (LON), we conduct a thorough analysis of two types of instances of the Quadratic Assignment Problem (QAP). This network model is a reduction of the landscape in which the nodes correspond to the local optima, and the edges account for the notion of adjacency between their basins of attraction. The model was inspired by the notion of ‘inherent network’ of potential energy surfaces proposed in physical-chemistry. The local optima networks extracted from the so called uniform and real-like QAP instances, show features clearly distinguishing these two types of instances. Apart from a clear confirmation that the search difficulty increases with the problem dimension, the analysis provides new confirming evidence explaining why the real-like instances are easier to solve exactly using heuristic search, while the uniform instances are easier to solve approximately. Although the local optima network model is still under development, we argue that it provides a novel view of combinatorial landscapes, opening up the possibilities for new analytical tools and understanding of problem difficulty in combinatorial optimization.

Local Optima Networks: A New Model of Combinatorial Fitness Landscapes

This chapter overviews a recently introduced network-based model of combinatorial landscapes: Local Optima Networks (LON). The model compresses the information given by the whole search space into a smaller mathematical object that is a graph having as vertices the local optima and as edges the possible weighted transitions between them. Two definitions of edges have been proposed: basin-transition and escape-edges, which capture relevant topological features of the underlying search spaces. This network model brings a new set of metrics to characterize the structure of combinatorial landscapes, those associated with the science of complex networks. These metrics are described, and results are presented of local optima network extraction and analysis for two selected combinatorial landscapes: NK landscapes and the quadratic assignment problem. Network features are found to correlate with and even predict the performance of heuristic search algorithms operating on these problems.

Local optima networks with escape edges

This paper proposes an alternative definition of edges (escape edges) for the recently introduced network-based model of combinatorial landscapes: Local Optima Networks (LON). The model compresses the information given by the whole search space into a smaller mathematical object that is the graph having as vertices the local optima and as edges the possible weighted transitions between them. The original definition of edges accounted for the notion of transitions between the basins of attraction of local optima. This definition, although informative, produced densely connected networks and required the exhaustive sampling of the basins of attraction. The alternative escape edges proposed here do not require a full computation of the basins. Instead, they account for the chances of escaping a local optima after a controlled mutation (e.g. 1 or 2 bit-flips) followed by hill-climbing. A statistical analysis comparing the two LON models for a set of NK landscapes, is presented and discussed. Moreover, a preliminary study is presented, which aims at validating the LON models as a tool for analyzing the dynamics of stochastic local search in combinatorial optimization.

The community structure of the European network of interlocking directorates 2005-2010.

The boards of directors at large European companies overlap with each other to a sizable extent both within and across national borders. This could have important economic, political and management consequences. In this work we study in detail the topological structure of the networks that arise from this phenomenon. Using a comprehensive information database, we reconstruct the implicit networks of shared directorates among the top 300 European firms in 2005 and 2010, and suggest a number of novel ways to explore the trans-nationality of such business elite networks. Powerful community detection heuristics indicate that geography still plays an important role: there exist clear communities and they have a distinct national character. Nonetheless, from 2005 to 2010 we observe a densification of the boards interlocks network and a larger transnational orientation in its communities. Together with central actors and assortativity analyses, we provide statistical evidence that, at the level of corporate governance, Europe is getting closer.

Local optima networks, landscape autocorrelation and heuristic search performance

Recent developments in fitness landscape analysis include the study of Local Optima Networks (LON) and applications of the Elementary Landscapes theory. This paper represents a first step at combining these two tools to explore their ability to forecast the performance of search algorithms. We base our analysis on the Quadratic Assignment Problem (QAP) and conduct a large statistical study over 600 generated instances of different types. Our results reveal interesting links between the network measures, the autocorrelation measures and the performance of heuristic search algorithms.

Optimizing the robustness of scale-free networks with simulated annealing

We study the robustness of Barabasi-Albert scale-free networks with respect to intentional attacks to highly connected nodes. Using the simulated annealing optimization heuristic, we rewire the networks such that their robustness to network fragmentation is improved but without changing neither the degree distribution nor the connectivity of single nodes. We show that simulated annealing improves on the results previously obtained with a simple hill-climbing procedure. We also introduce a local move operator in order to facilitate actual rewiring and show numerically that the results are almost equally good.

Generating Robust and Efficient Networks Under Targeted Attacks

Much of our commerce and travel depends on the efficient operation of large scale networks. Some of those, such as electric power grids, transportation systems, communication networks, and others, must maintain their efficiency even after several failures, or malicious attacks. We outline a procedure that modifies any given network to enhance its robustness, defined as the size of its largest connected component after a succession of attacks, whilst keeping a high efficiency, described in terms of the shortest paths among nodes. We also show that this generated set of networks is very similar to networks optimized for robustness in several aspects such as high assortativity and the presence of an onion-like structure.