%0 Journal Article
%T Traffic Signal Settings Optimization Using Gradient Descent
%A Możejko, Marcin
%A Brzeski, Maciej
%A Mądry, Łukasz
%A Skowronek, Łukasz
%A Gora, Paweł
%J Schedae Informaticae
%V 2018
%R 10.4467/20838476SI.18.002.10407
%N Volume 27
%P 19-30
%K traffic optimization, metamodels, activation functions, genetic algorithm, gradient descent
%@ 1732-3916
%D 2018
%U https://ejournals.eu/en/journal/schedae-informaticae/article/traffic-signal-settings-optimization-using-gradient-descent
%X 
	We investigate performance of a gradient descent optimization (GR) applied to the traffic signal setting problem and compare it to genetic algorithms. We used neural networks as metamodels evaluating quality of signal settings and discovered that both optimization methods produce similar results, e.g., in both cases the accuracy of neural networks close to local optima depends on an activation function (e.g., TANH activation makes optimization process converge to different minima than ReLU activation).