Retrieving actions from goal hierarchies

Reaching Into the Unknown: Actions, Goal Hierarchies, and Explorative Agency

Action is widely characterized as possessing a teleological dimension. The dominant way of describing goal-directed action and agency is in terms of exploitation, i.e., pursuing pre-specified goals using existing strategies. Recent theoretical developments emphasize the place of exploration, i.e., discovering new goals or acquiring new strategies. The exploitation-exploration distinction poses ...

Dynamic Goal Hierarchies

In this paper we introduce and formalise dynamic goal hierarchies. We begin with a formal definition of goals, according to which they are rational desires. In particular, we require that an agent’s goals are coherent; that is, that the agent believes that each goal is jointly realisable with all of the goals which the agent considers to be more important. Thus an agent’s goals form a hierarchy...

Retrieving Actions in Group Contexts

We develop methods for action retrieval from surveillance video using contextual feature representations. The novelty of our proposed approach is two-fold. First, we introduce a new feature representation called the action context (AC) descriptor. The AC descriptor encodes information about not only the action of an individual person in the video, but also the behaviour of other people nearby. ...

Goal-directed actions

Inductive Generalization of Analytically Learned Goal Hierarchies

We describe a new approach for learning procedural knowledge represented as Teleoreactive Logic Programs (TLPs) using relational behavior traces as input. A TLP organizes task decomposition skills hierarchically and attaches them explicitly defined goals. Our approach integrates analytical learning with inductive generalization in order to learn these skills. The analytical component predicts t...