Projects

Here, I experimented with the branch of machine and deep learning called 'meta-learning'. This is what is known as 'learning from learning', or using a machine learning agent to learn from the results of another machine learning agent. Here, I simulated the Cartpole OpenGym environment using a neural network, and learned from the results of those simulations using a Q-Learning agent. The second agent would control the hyperparameters of the first agent, from what it learns. Also provided is a link to the Google Colab document that contains all the code.

This app was somewhere in between a passion project, and a final project for a class. I had always wanted to make this app for fun, and found the perfect opportunity to do so in a university course. It is based off the subreddit called r/IncreasinglyVerbose. The idea is that you type in some text, and it will generate a much more verbose way to say the same phrase. It was a lot of fun! Take a look at the video to see the demonstration.

This project was focused on simulating the famous Birthday Paradox, in the Julia language. This paradox is very interesting to study, and I was able to prove it's truth in my own simulations. Watch the video to learn about the paradox and see the results of my simulation. I have also provided a link below to the repository which houses the code for my simulation.