/Robustness-Huber-vs-Hinge-Loss

Verified the robustness to outliers of Huber Loss with Hinge loss for classification over different datasets.

Primary LanguageJupyter Notebook

Readme

The main objective of this project is to show the comparison between Huber Loss (for classification) and Hinge Loss with respect to robustness to outliers. We have added 0%, 15%, and 30% Noise to Training and Validation set of the data and then checked the performance of our model. We have applied Linear SVM and SGD classifier model in this project.

This work is a part of my Summer Research Internship at IIT (BHU), Varanasi under Dr. K.K. Shukla, Professor, Computer Science Dept.

If you have any queries or suggestions you can mail me: vkasojhaa@gmail.com