Project Page
Welcome to my project page! This is where I will be link to all of my personal machine learning projects. I'm building this portfolio now (updated 5/23), and have lots of projects I need to write up and add to this page. As I upload projects old and new, this is where they will go.
Uploaded Projects
Dominion AI
One of my favorite games is Dominion, a game about building a deck of cards
that can efficiently earn points. In this project, I tackle the challenge
of building an AI to play the game (Updated 2/24).
Predator/prey simulation
Here I use reinforcement learning to create AI prey and AI predators, who
both attempt to survive and compete against each other.
Updated (5/23).
Fantasy Football Forecasting
This is a personal, ongiong project to predict the performance of players
in fantasy football. In it's current state, the program operates only on teams,
and not individual players, but does a good job predicting team performance.
Updated (11/22).
What2Lego?
This was a project done as part of my fellowship with the Insight program.
From start to finish, I developed a product that could identify a group of
Legos, and using them, recommend a set to build. For more detail
watch my presentation outlining the project.
Updated (6/15/18).
Deep Learning Number Recognition
Here I take a crack at that classic problem, recognizing images of numbers.
I use the MNIST dataset in a Kaggle competition, and focus on exploring the
differences between modeling with Scikit-Learn's Multi Layered Peceptron, and
TensorFlow.
Updated (5/31/18).
Monthly Cancellations for Pet Insurance
This was an interview project I was presented with. The goal was to
predict which customers cancel their policies in January 2017,
with limited customers and claims information. I explore both
machine learning and survival analysis to predict the lifetime of
customer accounts. Published with permission from the Trupanion.
Updated (11/14/17).
