Learning Type-Generalized Actions for Symbolic Planning
This addresses the need for more flexible and transferable symbolic planning methods for intelligent agents, though it appears incremental as it builds on existing symbolic planning with new generalization techniques.
The paper tackles the problem of hand-designed symbolic representations limiting transferability in symbolic planning by proposing type-generalized actions learned from few observations, showing they can generalize to novel situations and solve unseen task combinations in a simulated kitchen environment.
Symbolic planning is a powerful technique to solve complex tasks that require long sequences of actions and can equip an intelligent agent with complex behavior. The downside of this approach is the necessity for suitable symbolic representations describing the state of the environment as well as the actions that can change it. Traditionally such representations are carefully hand-designed by experts for distinct problem domains, which limits their transferability to different problems and environment complexities. In this paper, we propose a novel concept to generalize symbolic actions using a given entity hierarchy and observed similar behavior. In a simulated grid-based kitchen environment, we show that type-generalized actions can be learned from few observations and generalize to novel situations. Incorporating an additional on-the-fly generalization mechanism during planning, unseen task combinations, involving longer sequences, novel entities and unexpected environment behavior, can be solved.