Yasuo Sugai

Hierarchical algorithm for a partition problem using simulated annealing: application to placement in VLSI layout

A hierarchical simulated annealing (HSA) method for a partition problem is proposed for a realization of parallelization to reduce computational time. The HSA method hierarchically divides a given problem into some sub-problems applying the SA method at each stage and makes it possible automatically to choose suitable parameters in the cost functions at each stage of the hierarchical computation. Theoretical consideration and application of the HSA method to the placement problem in VLSI layout show that the HSA method can accomplish much more reduction of computational time and a better final result than the original SA method.

Method of flow-based modelling for large-scale systems: the basic idea

This paper proposes a method of flow-based modelling for large-scale systems that is based on the following two characteristics of the large-scale system: (a) multi-aspects and (b) the topological relations of the flows. Two (generally several) different kinds of flow play an important role in the modelling of a large-scale system. Under the assumption that one knows the topology of the flows (information about the origins and terminal points of all flows), it becomes possible to compensate for the lack of data about one aspect of the large-scale system by using the available information on another facet. The goal of the proposed method is not only to approximate the equations of systems behaviour, but also to make evident some properties and principles hidden in the system.

Generator of learning data for the TSPs based on the visiting order of the cities on convex hull

The optimal tours of the traveling salesman problems( TSPs) in two dimensional Euclidean space have the characteristics in the visiting order of the cities on the convex hull. Based on this characteristics, the TSPs can be replaced into some shortest Hamiltonian path problems(SHPPs) of which solutions assemble a tour. This reduction is enabled by the calculation of convex hull and the classification of the cities not on the convex hull into the subsets of cities which construct SHPPs. This procedure means that the TSPs are equivalent to the classification problems, which leads to be able to apply existing methods of machine learning to the TSPs. We show that the teaching data for machine learning are available as the optimal classifications in the instances of which the optimal tour has been found.

Spatial lattice habitat preference model: pattern formation of population

A mathematical model emphasizing the development of spatial heterogeneity and the autonomous behaviour of constituent species is described. This model, which is called the Spatial Lattice Habitat Preference Model (SLHP model), addresses the relationship between microscopic interactions among individuals within patches and macroscopic spatial patterns produced. Numerical calculations are used to show the various ways that spatial pattern formation occurs when environmental differences exist. Without environmental differences, populations become homogeneous in space and the macroscopic SLHP model becomes a Volterra system, as a special case. With environmental heterogeneity, however, population oscillations, when they occur, tend to be of relatively small magnitude. In as much as these behaviours accord with present experimental and field evidence, the SLHP model may be a promising tool for modelling the dynamics of spatially heterogeneous ecological systems.