optimization

Tools to explore the Potential Energy Surface (PES) for atomic and molecular systems and generate candidate structures for the local and global minima.

Machine Learning Basics - implementation of: loss functions (cross entropy loss, L1 loss, L2 loss, hinge loss), regularizations (L1 regularizer, L2 regularizer, early stopping), gradient check, optimizers, training a simple deep model.

Optimal portfolio selection in Python

Minimize the sum of pair distances for a system of particles N-space. This project is to approximate the smallest sum of pair distances among particles, where distances are calculated pairwise and every particle is the only counted once