Ekhiñe Irurozki -- PhD defense
Our group member Ekhiñe Irurozki will defend her PhD on October the 31st, friday, in the Computer Science Faculty, San Sebastian, 11:30AM. The title of the PhD is "Sampling and Learning Distance-based Probability Models for Permutation Spaces". PhD advisors are Jose A. Lozano and Borja Calvo.
A link to the PhD document can be found here.
New impact factor publication -- Weather and Forecasting
Our group members Pablo Rozas-Larraondo, Inaki Inza and Jose Antonio Lozano recently completed the following publication:
"A method for wind speed forecasting in airports based on non-parametric regression". Weather and Forecasting.
Wind is one of the parameters best predicted by numerical weather models, as it can be directly calculated from the physical equations of pressure which govern its movement. However, local winds are highly affected by topography which global numerical weather models, due to their limited resolution, are not able to reproduce. In order to improve the skill of numerical weather models, statistical and data analysis methods can be used. Machine learning techniques can be applied to train a model with data coming from both the model and observations in the area of interest. In this paper, a new method based on Non-parametric Multivariate Locally Weighted Regression is studied for improving the forecasted wind speed of a numerical weather model. Wind direction data is used to build different regression models, as a way of accounting for the effect of surrounding topography. The use of this technique offers similar levels of accuracy for wind speed forecasts compared with other machine learning algorithms with the advantage of being more intuitive and easy to interpret.
New impact factor publication -- European Journal of Operation Research
Our group members Josu Ceberio, Alexander Mendiburu and Jose Antonio Lozano recently completed the following publication:
"The linear ordering problem revisited". European Journal of Operation Research.
The Linear Ordering Problem is a popular combinatorial optimisation problem which has been extensively addressed in the literature. However, in spite of its popularity, little is known about the characteristics of this problem. This paper studies a procedure to extract static information from an instance of the problem, and proposes a method to incorporate the obtained knowledge in order to improve the performance of local search-based algorithms. The procedure introduced identifies the positions where the indexes cannot generate local optima for the insert neighbourhood, and thus global optima solutions. This information is then used to propose a restricted insert neighbourhood that discards the insert operations which move indexes to positions where optimal solutions are not generated. In order to measure the efficiency of the proposed restricted insert neighbourhood
system, two state-of-the-art algorithms for the LOP that include local search procedures have been modified. Conducted experiments confirm that the restricted versions of the algorithms outperform the classical designs systematically when a maximum number of function evaluations is considered as the stopping criterion. The statistical test included in the experimentation reports significant differences in all the cases, which validates the efficiency of our proposal. Moreover, additional experiments comparing the execution times reveal that the restricted approaches are faster than their counterparts for most of the instances.
New impact factor publication -- Transportmetrica A: Transport Science
Our group members Usue Mori, Alexander Mendiburu and Jose Antonio Lozano, together with Maite Alvarez, a researcher of the Tecnalia corporation, recently completed the following publication:
"A Review of Travel Time Estimation and Forecasting for Advanced Traveller Information Systems". Transportmetrica A: Transport Science
Due to the increase in vehicle transit and congestion in road networks, providing information about the state of the traffic to commuters has become a critical issue for Advanced Traveller Information Systems. These systems should assist users in making pre-trip
decisions and, for this purpose, delivering travel time information is very useful because it is very intuitive and easily understood by all travellers.
The aim of this paper is to present a global view of the literature on the modelling of travel time, introducing essential concepts and giving a thorough classification of the existing techniques. Most of the attention will focus on travel time estimation and travel time prediction, which are two of the most relevant challenges in travel time modelling. The definition and goals of these two modelling tasks along with the methodologies used to carry them out will be further explored and categorized.
New impact factor publication -- Information Sciences
Our group member Alexander Mendiburu, together with a group of colleagues of the Polytechnic University of Madrid, recently completed the following publication:
"Distributed estimation of distribution algorithms for continuous optimization: how does the exchanged information influence their behavior". Information Sciences.
One of the most promising areas in which probabilistic graphical models have shown an incipient activity is the field of heuristic optimization and, in particular, in Estimation of Distribution Algorithms. Due to their inherent parallelism, different research lines have been studied trying to improve Estimation of Distribution Algorithms from the point of view of execution time and/or accuracy. Among these proposals, we focus on the so-called distributed or island-based models. This approach defines several islands (algorithms instances) running independently and exchanging information with a given frequency. The information sent by the islands can be either a set of individuals or a probabilistic model. This paper presents a comparative study for a distributed univariate Estimation of Distribution Algorithm and a multivariate version, paying special attention to the comparison of two alternative methods for exchanging information, over a wide set of parameters and problems – the standard benchmark developed for the IEEE Workshop on Evolutionary Algorithms and other Metaheuristics for Continuous Optimization Problems
of the ISDA 2009 Conference. Several analyses from different points of view have been conducted to analyze both the influence of the parameters and the relationships between them including a characterization of the configurations according to their behavior on the proposed benchmark.
New impact factor publication -- IEEE Transactions on Parallel and Distributed Systems
Our group members Unai Lopez-Novoa, Alexander Mendiburu and Jose Miguel-Alonso, recently completed the following publication:
"A survey of performance modeling and simulation techniques for accelerator-based computing". IEEE Transactions on Parallel and Distributed Systems.
The high performance computing landscape is shifting from collections of homogeneous nodes towards heterogeneous systems, in which nodes consist of a combination of traditional out-of-order execution cores and accelerator devices. Accelerators, built around GPUs, many-core chips, FPGAs or DSPs, are used to offload compute-intensive tasks. The advent of this type of systems has brought about a wide and diverse ecosystem of development platforms, optimization tools and performance analysis frameworks. This is a review of the state-of-the-art in performance tools for heterogeneous computing, focusing on the most popular families of accelerators: GPUs and Intel’s Xeon Phi. We describe current heterogeneous systems and the development frameworks and tools that can be used for developing for them. The core of this survey is a review of the performance models and tools, including simulators, proposed in the literature for these platforms.
Page 1 of 5