Projects/Hobbies
Click to be pointed to Github Repo
Reinforcement Learning: Lego Car
In this project, a major achievement was reached as I was able to train a RL agent in a simulated environment and transfer the agent to a lego robot using a Raspberry Pi. The goal of the agent is to simply turn left or right when following a blue line.
FRBNY - TIme Series Analysis
This project was a bit more of a test in terms of how good I am with Python. Essentially, I converted a nowcast model from MatLab to Python that the Federal Reserve uses. A bit of a challenge but got it done!
My hello world to Reinforcement Learning
Reinforcement Learning is an interesting paradigm in terms of achieving/optimizing a process, especially when you may not have data and/or don't know what the optimal solution looks like. In this project, we create a simple agent that learns to avoid green bars. Super simple problem but helps to understand how RL works mechanically.
Simulated Agent!!!
The purpose of this project is to figure out how to train an RL agent using Coppeliasim! The idea is to train an agent on a simulated environment, and then transfer that to the real world