Projects

A list of my machine learning and other projects that might interest you

  • Web Scraping and URL Classification API

    A Web Scraping and URL Classification API that extracts content from URLs and applies a machine learning model to classify them based on their text content. Supports batch processing of URLs, provides confidence scores, and returns the top three predicted categories for each URL. Learn more

  • Game of Life - QT

    An interactive implementation of Conway's Game of Life built with Qt. Features an adjustable grid size, interactive controls for starting, stopping, and resetting the simulation, and intuitive cell state toggling with wrapping edges for continuous simulation. Learn more

  • Reinforcement Learning with Unity ML-Agents

    A project exploring reinforcement learning using Unity ML-Agents in five environments: Basic, 3D Balance Ball, Grid World, Push Block, and Worm. Utilized RL algorithms such as DQN, DDPG, and SAC to train agents for tasks like navigation, balance, and object manipulation. Highlighted the effectiveness of Unity ML-Agents in creating complex training scenarios for diverse tasks. Learn more

  • Lunar Lander Using DDPG and SAC

    An implementation of DDPG and SAC for the OpenAI Gym Lunar Lander environment. Utilizes PyTorch, OpenAI Gym, and Tensorboard to train and visualize reinforcement learning agents. Demonstrates the performance of the agents through episodic returns during training. Learn more

  • CartPole Balancing Using DQN

    An implementation of a Deep Q-Learning (DQL) network for the OpenAI Gym CartPole-v1 environment. Utilizes PyTorch, OpenAI Gym, and Tensorboard to train and visualize the agent's performance. Shows training progress through episodic returns with different smoothing levels. Learn more

  • CARLA: V2V LiDAR Fusion

    A collaborative vehicle-to-vehicle (V2V) communication framework leveraging LiDAR data sharing among smart vehicles to enhance autonomous navigation. Integrates LiDAR data into the ego vehicle's perception pipeline to detect occluded and long-range actors, improving safety and trajectory planning. Demonstrates the effectiveness of V2V communication for real-time collaborative perception in complex urban environments. Learn more

  • Deep Learning Network Intrusion Detection

    A deep learning-based Intrusion Detection System (IDS) utilizing a lightweight and compact feature vector set. Implements a Feed-Forward Neural Network (FFNN) to enhance detection accuracy for network threats while reducing computational overhead. Achieved superior accuracy on UNSW-NB15 and NSL-KDD datasets compared to traditional approaches. Learn more