Reinforcement Learning exhibit for the I AM AI exhibition

by IMAGINARY - ai.imaginary.org

Build:

• Use webpack to produce a single js file, through the dist task (npm run dist)
• Use sass to compile the scss into css (npm run sass or npm run watch:sass for starting a process that watches on the css and recompiles it as soon as modified)

Source files in src. Output folder is dist.

Debug mode

Press the D key to show the maximum Q value from all possible actions in each state (cell).