Generalized Planning With Deep Reinforcement Learning
This addresses the challenge of scalable and efficient planning in AI, offering a method to deduce general principles from small examples for broader application, though it is incremental in combining existing techniques.
The paper tackles the problem of learning generalized policies for planning domains using Deep Reinforcement Learning and Graph Neural Networks, demonstrating that these policies can generalize to instances orders of magnitude larger than those used in training.
A hallmark of intelligence is the ability to deduce general principles from examples, which are correct beyond the range of those observed. Generalized Planning deals with finding such principles for a class of planning problems, so that principles discovered using small instances of a domain can be used to solve much larger instances of the same domain. In this work we study the use of Deep Reinforcement Learning and Graph Neural Networks to learn such generalized policies and demonstrate that they can generalize to instances that are orders of magnitude larger than those they were trained on.