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

Figure 4.

Figure 4

State-action space coverage during early learning. The policy based on Artificial Curiosity (AC) explores the state-action space most efficiently, compared to policies based on random exploration (RAND) and always selecting the least tried state-action (LT). Time is measured in state transitions.