Tyler Kinkade
This was a small exploratory project to compare the effectiveness of artificial agent algorithms ranging in sophistication from naive to deep reinforcement learning (Russell & Norvig, 2022; Sutton & Barto, 2018) to successfully land a virtual "lunar lander" in a Gymnasium (2022) model environment. After comparing the various models, I explored the effects of various hyperparameters on a deep Q-network model (Mnih et al., 2015).
See the project.ipynb Jupyter notebook for the full report.