Graph RL lets you visualize the training process of a reinforcement learning agent trying to find the shortest path in a graph.
At first, the agent uses random choices to try to explore the graph As the Agent lands on a Node, it gets rewarded based on the node's probability of being in the shortest path. It then calculates a q-value for every node in the graph.
The shortest path consists of the nodes with the best q-values
How to play?
1- Click on two nodes to select a starting and ending point
2- Choose the number of iterations
3- Train it
Legend
A
Node the Agent is currently on
B
Neighbors of the current Node
Edge is in shortest path
Press enter or space to select a node.You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel.
Press enter or space to select an edge. You can then press delete to remove it or escape to cancel.