Projects done in the AI learning algorithm class @ UCSD

Code for the paper "Block-coordinate primal-dual algorithm for linearly constrained optimization problem"

Sample Convex Optimization using Gradient Descent, Newton's method and Coordinate Descent

Implementation of optimization algorithms in python

CycleSL: Server-Client Cyclical Update Driven Scalable Split Learning

Predicting goodness points of a wine given its reviews

Implementation of Relaxed Lasso Algorithm for Linear Regression.

Implementations of various numerical optimization methods, written in plain Java.

Workshop on the course "Computer decision support systems" at V. N. Karazin Kharkiv National University

The implementation of Coordinate Descent Method Accelerated by Universal Metaalgorithm with efficient amortised complexity of iteration & Experiments with sparse SoftMax function, where the proposed method is better than FGM

Hello! All codes belong to me. I created those codes for my Machine Learning Lab Class. Enjoy it!

The lasso function (from scratch) to estimate parameters for an nxp matrix when p >> n

Implementation of algorithms for stereo vision

Powell's example of cyclic non-convergence in coordinate descent with exact minimization

Coordinate descent

Solving Quadratic Programming problem with LASSO constraint

This repository contains codes and functions for Ridge Regression (Normal Eqquation method and Coordinate Descent method) and Lasso Regression (Coordinate Descent method). There is some analysis of the preformance of these funcions/models. There is also a comparison of these with the sklearn.

Block-coordinate Gauss-Newton

Library of Semi-Relaxed Optimal Transport