Deep learning of support vector machines with class probability output networks

In the article the analysis of different approaches to invariant target recognition was made, such as based on the support vector machines, deep learning techniques, neural networks, generation of moment features, etc. It was determined that one of the perspectives approaches in target recognition suppose the use of the central and logarithmic central image chords transformations. There have been described the new methods of the target recognition, based on the central image chords transformation. Tasks of target recognition were formulated. New 4 methods of target recognition were described. It is presented the comparison of the different target’s recognition methods regarding the processing stages number, realized operations, target’s image normalization’s operation, the operations realized in parallel, kind of the target’s scale and rotation determination sequence, target’s rotation determination approach.

Download Full-text

Similarity-Based Summarization of Music Files for Support Vector Machines

Complexity ◽

10.1155/2018/1935938 ◽

2018 ◽

Vol 2018 ◽

pp. 1-10 ◽

Cited By ~ 2

Author(s):

Jan Jakubik ◽

Halina Kwaśnicka

Keyword(s):

Neural Network ◽

Deep Learning ◽

Support Vector Machines ◽

Loss Function ◽

Expert Knowledge ◽

Feature Learning ◽

Compact Representation ◽

Support Vector ◽

Learning Methods ◽

Vector Machines

Automatic retrieval of music information is an active area of research in which problems such as automatically assigning genres or descriptors of emotional content to music emerge. Recent advancements in the area rely on the use of deep learning, which allows researchers to operate on a low-level description of the music. Deep neural network architectures can learn to build feature representations that summarize music files from data itself, rather than expert knowledge. In this paper, a novel approach to applying feature learning in combination with support vector machines to musical data is presented. A spectrogram of the music file, which is too complex to be processed by SVM, is first reduced to a compact representation by a recurrent neural network. An adjustment to loss function of the network is proposed so that the network learns to build a representation space that replicates a certain notion of similarity between annotations, rather than to explicitly make predictions. We evaluate the approach on five datasets, focusing on emotion recognition and complementing it with genre classification. In experiments, the proposed loss function adjustment is shown to improve results in classification and regression tasks, but only when the learned similarity notion corresponds to a kernel function employed within the SVM. These results suggest that adjusting deep learning methods to build data representations that target a specific classifier or regressor can open up new perspectives for the use of standard machine learning methods in music domain.

Download Full-text