So far, we have finished explaining what minesweeper is to the audience using scrollytelling. It explains the rules of minesweeper, scenarios within the game, and what each icon means. Additionally, we created a layout for how our webpage will flow with different frames showing each visualization. We also found projects similar to ours, which is Reinforcement Learning with Minesweeper, and will improve those projects through more interactive visualizations.
The most challenging part of the project will be implementing the Reinforcement Learning for Minesweeper, because we will have to keep track of many different variables to visualize throughout the process. One difficult visualization would be comparing the initial model to the final model by showing one game state and using a slider to go across the moves. This will be difficult to code because we have to code out the different game states at each move. We will have to explain Deep Q-learning in the context of minesweeper, which could prove difficult as well. Afterall, deep learning is a difficult subject to explain to people with no background in machine learning
Acknowledgements: