Wavelet techniques are effective for obtaining data representations or features, which you can use in machine learning and deep learning workflows.
Wavelet scattering enables you to produce low-variance data representations, which are invariant to translations on a scale you define and are continuous with respect to deformations. Wavelet scattering requires few user-specified parameters to produce compact representations of data. You can use these representations in conjunction with machine learning algorithms for classification and regression.
You can use the continuous wavelet transform (CWT) to generate 2-D time-frequency maps of time series data, which can be used as image inputs with deep convolutional neural networks (CNN). Generating time-frequency representations for use in deep CNNs is a powerful approach for signal classification. The ability of the CWT to simultaneously capture steady-state and transient behavior in time series data makes the wavelet-based time-frequency representation particularly robust when paired with deep CNNs.
Wavelet methods can also be used to generate sparse feature vectors for statistical learning applications. The sparsity property of wavelet representations enables you to achieve significant dimensionality reduction without sacrificing discriminability.
Derive low-variance features from real-valued time series and image data.
This example shows how changing the invariance scale and oversampling factor affects the output of the wavelet scattering transform.
Pretrained Deep Neural Networks (Deep Learning Toolbox)
Learn how to download and use pretrained convolutional neural networks for classification, transfer learning and feature extraction.
Train Support Vector Machines Using Classification Learner App (Statistics and Machine Learning Toolbox)
Create and compare support vector machine (SVM) classifiers, and export trained models to make predictions for new data.