Master's Thesis

Recently, machine learning techniques have been put into practice in precisely characterizing various dynamical properties or phenomena. Here we make use of supervised machine learning algorithms for the model-free prediction of factors determining or controlling the intensity of symme- try breaking phenomena emergent in different network architectures. In an attempt to achieve this, chimera states (solitary states) are engineered by establishing delays in the neighboring links of a node (the interlayer links) in a 2-D lattice (multiplex network) of oscillators. Different machine learning classifiers, K-Nearest Neighbours (Knn), Support Vector Machine (SVM) and Multi-Layer Perceptron Neural Network (MLP-NN) are then employed, feeding on the data obtained from mentioned models, for the prediction of intensity of rippling chimera states and critical delay to char- acterize solitary states. It is revealed from our analysis that Multi-Layer Perceptron Neural Network (MLP-NN) classifier is best suited for the char- acterization of the engineered chimera and solitary states. We hope that our successful attempt in characterizing a class of partially synchronized states using machine learning techniques would be useful in broadening the scope of model-free machine learning techniques in characterizing other phase states as well.

Paper↗