Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron... more
Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron (MLP). However, it suffers from some drawbacks such as getting trapped in local minima, human intervention during the stage of training, and limitations in architecture design. The aims of this study were twofold. The first was to employ NeuroEvolution of Augmenting Topologies (NEAT), which has many successful applications in numerous fields. In this paper, we applied it to time series forecasting for the first time and compared its performance with that of the MLP. The second aim was to analyse the performance resulting from the pairwise combination of these methods. In general, the results suggested that the forecasts from the NEAT algorithm were more accurate than those of the MLP. The results also showed that pairwise combined forecasts in gene...
There are two important limitations of standard tree-based genetic programming (GP). First, GP tends to evolve unnecessarily large programs, what is referred to as bloat. Second, GP uses inefficient search operators that focus on... more
There are two important limitations of standard tree-based genetic programming (GP). First, GP tends to evolve unnecessarily large programs, what is referred to as bloat. Second, GP uses inefficient search operators that focus on modifying program syntax. The first problem has been studied extensively, with many works proposing bloat control methods. Regarding the second problem, one approach is to use alternative search operators, for instance geometric semantic operators , to improve convergence. In this work, our goal is to experimentally show that both problems can be effectively addressed by incorporating a local search op-timizer as an additional search operator. Using real-world problems, we show that this rather simple strategy can improve the convergence and performance of tree-based GP, while also reducing program size. Given these results, a question arises: why are local search strategies so uncommon in GP? A small survey of popular GP libraries suggests to us that local search is underused in GP systems. We conclude by outlining plausible answers for this question and highlighting future work.
Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron... more
Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron (MLP). However, it suffers from some drawbacks such as getting trapped in local minima, human intervention during the stage of training, and limitations in architecture design. The aims of this study were twofold. The first was to employ NeuroEvolution of Augmenting Topologies (NEAT), which has many successful applications in numerous fields. In this paper, we applied it to time series forecasting for the first time and compared its performance with that of the MLP. The second aim was to analyse the performance resulting from the pairwise combination of these methods. In general, the results suggested that the forecasts from the NEAT algorithm were more accurate than those of the MLP. The results also showed that pairwise combined forecasts in general were better than single forecasts. The best forecasts of all were obtained by pairwise combination of MLP and NEAT.
Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron... more
Neural networks are one of the widely-used time series forecasting methods in time series applications. Among different neural network architectures and learning algorithms, the most popular choice is the feedforward Multilayer Perceptron (MLP). However, it suffers from some drawbacks such as getting trapped in local minima, human intervention during the stage of training, and limitations in architecture design. The aims of this study were twofold. The first was to employ NeuroEvolution of Augmenting Topologies (NEAT), which has many successful applications in numerous fields. In this paper, we applied it to time series forecasting for the first time and compared its performance with that of the MLP. The second aim was to analyse the performance resulting from the pairwise combination of these methods. In general, the results suggested that the forecasts from the NEAT algorithm were more accurate than those of the MLP. The results also showed that pairwise combined forecasts in gene...
Die Geschichte der kürzlich eröffneten Neuen Eisenbahn-Alpentransversale (NEAT) - der beiden Alpenbahnlinien durch den Lötschbergtunnel und den Gotthardtunnel - geht auf Pläne der späten 1960er und frühen 1970er-Jahre zurück, die... more
Die Geschichte der kürzlich eröffneten Neuen Eisenbahn-Alpentransversale (NEAT) - der beiden Alpenbahnlinien durch den Lötschbergtunnel und den Gotthardtunnel - geht auf Pläne der späten 1960er und frühen 1970er-Jahre zurück, die Kompetivität und Attraktivität des Eisenbahnverkehrs gegenüber der Strasse durch Schnellbahnen und transnationale Verbindungen zu steigern. Allerdings wurde eine neue Alpenbahnlinie, die die 1882 eröffneten Gotthardbahn-Berglinie ersetzen könnte und von der Planer und Ingenieure schon 1947 träumten, bald dem vorrangigen nationalen und internationalen Interesse an einem Strassentunnel durch den Gotthard vorerst verdrängt. Ab Mitte der 1980er-Jahre rückte das Projekt aus zwei Gründen wieder auf die Agenda der Verkehrspolitik und der Bahnunternehmen: Erstens verlangten die Süd- und Alpenkantone der Schweiz bessere Anschlüsse, als das Parlament das Projekt für eine "Bahn 2000" diskutierte, das vor allem einen Bahnausbau im Mittelland bringen würde, diskutierte. Zweitens drängte die Europäische Gemeinschaft/Union im Hinblick auf den Gemeinsamen Markt auf die Ermöglichung des transalpinen Güterschwerverkehrs. Die NEAT wurde somit zur Verlängerung der Bahn 2000 und, als schnelle Auch-Güterbahn für den Huckepack-Verkehr konzipiert, zur Antwort auf die Anforderungen der europäischen Verkehrspolitik, die mit schweizerischen Bedürfnissen, den Alpenraum vor mehr Strassenverkehr zu schützen (Alpenschutz-Initiative) kollidierte.