Trained linear regression models using the least squares method to explore the trade-off between overfitting and underfitting.
Implemented the k-Nearest Neighbours (k-NN) algorithm and evaluated model performance using cross-validation techniques.
Applied k-means clustering for image segmentation and compression to reduce storage size while preserving visual quality.
Built a neural network from scratch with backpropagation training, then reimplemented the model using PyTorch for comparison.
Developed and trained reinforcement learning agents in the Lunar Lander environment using Q-Learning and Deep Q-Networks (DQN).
Demo:
