Learning sparse relational transition models
This work addresses the challenge of efficient model learning in robotics and AI planning, though it appears incremental as it builds on existing relational and neural methods.
The paper tackled the problem of learning transition models in complex uncertain domains by using relational rules and neural networks, achieving more versatility and sample efficiency than monolithic models in a simulated robot pushing task.
We present a representation for describing transition models in complex uncertain domains using relational rules. For any action, a rule selects a set of relevant objects and computes a distribution over properties of just those objects in the resulting state given their properties in the previous state. An iterative greedy algorithm is used to construct a set of deictic references that determine which objects are relevant in any given state. Feed-forward neural networks are used to learn the transition distribution on the relevant objects' properties. This strategy is demonstrated to be both more versatile and more sample efficient than learning a monolithic transition model in a simulated domain in which a robot pushes stacks of objects on a cluttered table.