Information Retrieval in High Dimensional Data (class deliverables)

PyTorch implementation of "Kernel Neural Optimal Transport" (ICLR 2023)

Comparison of principal components analysis with diffusion maps on toy data sets and a molecular simulation trajectory

Multi-Shot Approximation of Discounted Cost MDPs

To deal with non-linearly separable we use SVM's Kernel Trick which maps data to higher dimension!

A perceptron optimized with Numba that makes use of dual coordinates and kernel functions instead of classical dot product

University of Washington: CSE 446 (WIN '17) Machine Learning

This repository contains the projects and homework from CSC 591 Graph Data Mining

packer identification tool using SVM

This repository contains my projects in the context of the "Computational Intelligence" course (held at FUM).

My first steps to becoming AI engineer :)

In this project I developed a program that generates nonlinearly separable data. This data is mapped into a higher dimension using a kernel trick where the data becomes linearly separable

Exploring meachine learning techniques and algorithms. Including clustering algorithms, perceptron and, more.

Machine Learning Algorithms

simple classifier with perceptron (university assignment)

Understanding of Support Vector Machines (SVMs) and their implementations. Covers linear and non-linear SVM, also explains how to use the "kernel trick" to handle non-linearly separable data.

A multiclass Support Vector Machine that supports kernel tricks has been implemented by solving the SVM quadratic programming problem.

This is a homework about supervised learning and kernel trick for machine learning course @ FUM.

Implemented a multi-class SVM that supports multi-core parallel computing