Wavelet Clustering and Classification

• Grant number: 25/08017-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2025
• Status: Discontinued
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Aluísio de Souza Pinheiro
• Grantee: Lívia Yumi de Souza Tuzita Yamamoto
• Host Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 23/02538-0 - Time Series, Wavelets, High Dimensional Data and Applications, AP.TEM
• Associated scholarship(s): 25/21355-9 - Google DeepMind's WaveNet model: Wavelets and Learnable Filters, BE.EP.IC

Abstract

Clustering and classification methods are routinely used in machine learning. This application on a massive scale of techniques devised for low dimensional data may yield mediocre results. These scalability issues are true for iid data, time series and images. Some of the classical methods include K-th nearet neighbors, K-means, support vector machines, trees and florests, Schur measures, neural networks etc. This project proposes using wavelets, more specifically Discrete Wavelet transforms to represent the original data prior to clustering and/or classifying. For deep learning (SVM and NN), we will use wavelet kernels and filters instead of the usual kernels and filters, respectively.