TechPrufer
Get started
Two Sigma
Two Sigma logoTwo Sigma · Feature trail

Machine Learning Bot Development for Halite

1 post · Feb 2018

What this trail covers

Two Sigma has developed capabilities to build and support machine learning bots for AI programming challenges like Halite. This involves providing ML starter kits, exploring different ML approaches (supervised learning vs. reinforcement learning), and establishing best practices for developing and evaluating ML bots in complex game environments. The focus is on practical application of ML, efficient data utilization, and robust performance evaluation.

All Two Sigma trails
  1. Best Practices from Building a Machine Learning Bot for HaliteThis post details the development of a machine learning bot for the Halite AI programming challenge. It outlines best practices learned, including choosing appropriate ML approaches (supervised learning over reinforcement learning due to computational constraints), starting with a simplified problem, determining optimal data sample sizes for training, ensuring models evaluate relevant variables to avoid overfitting, and establishing effective performance evaluation metrics by comparing ML bots against each other and baseline logic. It also touches on catching edge cases through controlled testing environments.