A computationally efficient implementation of Kernel Principal Component Analysis (KPCA) using the Gaussian (RBF) kernel. This package leverages Rcpp and the Armadillo C++ library for fast matrix operations.

[NeurIPS 2024] Kernel PCA for Out-of-Distribution Detection

Re-Implementation of Gaussian Process Latent Variable Model algorithm & performance assessment against Kernel-PCA

Python package for plug and play dimensionality reduction techniques and data visualization in 2D or 3D.

This repository implements customer segmentation techniques to analyze credit card user behavior and identify distinct customer groups. By leveraging Python libraries like pandas, Scipy and scikit-learn.

Houses a series of projects I worked on for a course in Data Mining that I took in my Ph.D. Data Science program at UTEP in the Fall of 2022. Covers areas such as Regularized Logistic Regression, Optimization, Kernel Methods, PageRank, Kernel PCA, Association Rule Mining, Anomaly Detection, Parametric/Nonparametric Nonlinear Regression, etc.

This repository explores the interplay between dimensionality reduction techniques and classification algorithms in the realm of breast cancer diagnosis. Leveraging the Breast Cancer Wisconsin dataset, it assesses the impact of various methods, including PCA, Kernel PCA, LLE, UMAP, and Supervised UMAP, on the performance of a Decision Tree.

This repository is dedicated to the lab activities of the course of Unsupervised Learning @Units

Winning one of the DACON competition

Repository for the code of the "Introduction to Machine Learning" (IML) lecture at the "Learning & Adaptive Systems Group" at ETH Zurich.

The code for Principal Component Analysis (PCA), dual PCA, Kernel PCA, Supervised PCA (SPCA), dual SPCA, and Kernel SPCA

Applying NLP methods and kernel PCA on news dataset to build a clustering model

Machine Learning assignments from coursework.

This repository contains the Python code my blog post Image denoising techniques: A comparison of PCA, kernel PCA, autoencoder, and CNN. See post for more details and results.

5th semester project concerning feature engineering and nonlinear dimensionality reduction in particular.

UML dimensionality reduction and clustering models for predicting if a banknote is genuine or not based on the dataset from OpenML containing wavelet analysis results for genuine and forged banknotes - practical exercise. (Python 3)

Complete Tutorial Guide with Code for learning ML

Project on Non-Linear Dimensionality Reduction - ENSAE ParisTech

My notes for Prof. Klaus Obermayer's "Machine Intelligence 2 - Unsupervised Learning" course at the TU Berlin

Front-end speech processing aims at extracting proper features from short- term segments of a speech utterance, known as frames. It is a pre-requisite step toward any pattern recognition problem employing speech or audio (e.g., music). Here, we are interesting in voice disorder classification. That is, to develop two-class classifiers, which can…