Alexandre Linhares

A multiple-population evolutionary approach to gate matrix layout

This paper deals with a Very-Large-Scale Integrated systems design problem that belongs to the NP(Nondeterministic Polynomial)-hard class. The Gate Matrix Layout problem has numerous applications in the chip-manufacturing industry and in other industrial settings. A memetic algorithm is employed to solve a set of benchmark instances, and numerical comparisons with a highly competitive method—a microcanonical optimization approach—are performed. Beyond the effectiveness of the method, shown by the results obtained for these instances, an additional goal of this work is to study how the performance of the algorithm is affected by the use of multiple populations and of different individual-migration policies between such populations. The results signal a strong performance improvement of multiple populations over single population approaches. Finally, the proposed algorithm presents several refinements, like structured populations and a specially tailored local search.

Entanglement of perception and reasoning in the combinatorial game of chess: Differential errors of strategic reconstruction

We question Chase and Simons (1973) study concerning the content of the chess chunks, and we conduct a new variation of the classic chess reconstruction experiments, analyzing 25 types of possible reconstruction errors of grandmasters, masters, and beginners. The differences between the errors conducted in poor, intermediate, and strategically perfect reconstructions provide insights concerning the encoding of experts. The results obtained shed clear light into the debate concerning the importance of abstract thought (i.e., forward search) vs. perceptual processes (i.e., pattern recognition). We claim that a clear solution to this debate is ultimately unfeasible, as our experiments demonstrate high entanglement of perception and reasoning. Our results provide additional evidence that analogy is central to strategic thought in chess.

The Emergence of Miller's Magic Number on a Sparse Distributed Memory

Human memory is limited in the number of items held in ones mind—a limit known as “Millers magic number”. We study the emergence of such limits as a result of the statistics of large bitvectors used to represent items in memory, given two postulates: i) the Sparse Distributed Memory; and ii) chunking through averaging. Potential implications for theoretical neuroscience are discussed.

How Can Experts See the Invisible? Reply to Bilalić and Gobet

Experts in all fields are able to see what is invisible to others. Experts are also able to see what is visible to all-and this is explored by Bilalić and Gobet. We question the method of normalizing all subjects in an experimental condition, and asking experts to behave as if they were novices. We claim that method leads Bilalić and Gobet to a nonsequitur.

Leilões eletrônicos reversos multiatributo: uma abordagem de decisão multicritério aplicada às compras públicas brasileiras

Resumen pt: Leiloes sao instituicoes seculares utilizadas nas relacoes comerciais entre individuos e organizacoes. Proveem maior flexibilidade aos processos de dete...

Sparse distributed memory: understanding the speed and robustness of expert memory.

How can experts, sometimes in exacting detail, almost immediately and very precisely recall memory items from a vast repertoire? The problem in which we will be interested concerns models of theoretical neuroscience that could explain the speed and robustness of an experts recollection. The approach is based on Sparse Distributed Memory, which has been shown to be plausible, both in a neuroscientific and in a psychological manner, in a number of ways. A crucial characteristic concerns the limits of human recollection, the “tip-of-tongue” memory event—which is found at a non-linearity in the model. We expand the theoretical framework, deriving an optimization formula to solve this non-linearity. Numerical results demonstrate how the higher frequency of rehearsal, through work or study, immediately increases the robustness and speed associated with expert memory.

VLSI design: gate matrix layout problem

With applications ranging from fields as distinct as fuzzy modeling (Xiong 2001), autonomous robot behavior (Luk et al. 2001), learning with backpropagation (Foo et al. 1999), and multicriteria optimization (Viennet et al. 1996), evolutionary methods have become an indispensable tool for systems scientists. Although already studied in the past, an interesting emerging issue is the use of multiple populations, which is gaining increased momentum from the conjunction of two technologies. On the hardware side, computer networks, multi-processor computers and distributed processing systems (such as workstations clusters) are increasingly becoming widespread. Regarding the software issues, the introduction of Parallel Virtual Machine1 (PVM), and later the Message Passing Interface Standard2 (MPI), as well as web-enabled, object-oriented languages (such as Java3) have also had their role. Most Evolutionary Algorithms (EAs) are methods that are easy to parallelize as well as naturally suitable for heterogeneous systems. For most EAs the distribution of the tasks is relatively easy for most applications. The workload can be distributed at an individual or a population level; the final choice depending on the complexity of the computations involved.