RANSAC Template Library

Solve many kinds of least-squares and matrix-recovery problems

Robust estimations from distribution structures: Mean.

Robust Gaussian Process with Iterative Trimming

MATLAB implementation of "Provable Dynamic Robust PCA or Robust Subspace tracking", IEEE Transactions on Information Theory, 2019.

Robust estimations from distribution structures: Central moments.

This is an open source library that can be used to autofocus telescopes. It uses a novel algorithm based on robust statistics. For a preprint, see https://arxiv.org/abs/2201.12466 .The library is currently used in Astro Photography tool (APT) https://www.astrophotography.app/

Scikit learn compatible constrained and robust polynomial regression in Python

This is the implementation of the five regression methods Least Square (LS), Regularized Least Square (RLS), LASSO, Robust Regression (RR) and Bayesian Regression (BR).

ML Coursework focused on solving Computational Finance and Risk Assessment models

R Package implementing the Penalized Elastic Net S- and MM-Estimator for Linear Regression

Python implementation of RANSAC algorithm

regression algorithm implementaion from scratch with python (least-squares, regularized LS, L1-regularized LS, robust regression)

Code accompanying the paper "Globally Optimal Learning for Structured Elliptical Losses", published at NeurIPS 2019

In this repository, using the statistical software R, are been analyzed robust techniques to estimate multivariate linear regression in presence of outliers, using the Bootstrap, a simulation method where the construction of sample distribution of given statistics occurring through resampling the same observed sample.

In this project I have implemented 15 different types of regression algorithms including Linear Regression, KNN Regressor, Decision Tree Regressor, RandomForest Regressor, XGBoost, CatBoost., LightGBM, etc. Along with it I have also performed Hyper Paramter Optimization & Cross Validation.

Applied analysis on the Bayesian student-t "Robust" regression model with Jeffrey's prior. Compared its model performance and robustness of posterior distributions with the Gaussian model when outliers are present.

Regression for Boston Housing price prediction: Linear, Multiple, Robust, OLS, Regularization (Ridge-l1 norm, LASSO-l2 norm, ElasticNet)

A collection of projects completed in STAT courses.

2021 Fall term, CSE 701 Project 03