Skip to main content
. 2014 Jan 6;7:25. doi: 10.3389/fnbot.2013.00025

Figure 6.

Figure 6

Decay of intrinsic reward over time. These snapshots of the reward distribution state-actions (x-axis) over time (from top to bottom) show how our curious agent becomes bored as it builds a better and better model of the state-action space. Time is measured in state transitions.