Characteristics of the Feature-Model-Free system.

<p>(<b>A</b>) Focusing on a particular feature. The Feature-Model-Free system relies on a value function based on features. Choosing an action (e.g. <i>goL</i>, <i>goM</i> or <i>exp</i>), defines the feature it is focusing on (e.g. <i>L</i>ever, <i>M</i>agazine or nothing ). Once the action is chosen (e.g. <i>goM</i> in blue), only the value of the focused feature (e.g. ) is updated by a standard reward prediction error, while leaving the values of the other features unchanged. (<b>B</b>) Feature-values permit generalization. At a different place and time in the episode, the agent can choose an action (e.g. <i>goM</i> in blue) focusing on a feature (e.g. <i>M</i>) that might have already been focused on. This leads to the revision of the same value (e.g. ) for two different states (e.g. and ). Values of features are shared amongst multiple states.