Injecting Prior Knowledge for Transfer Learning into Reinforcement Learning Algorithms using Logic Tensor Networks
This work addresses the challenge of improving transfer learning in Reinforcement Learning by incorporating symbolic priors, though it appears incremental as it builds on existing methods with a proof-of-concept in simple settings.
The paper tackled the problem of enabling Reinforcement Learning agents to leverage prior knowledge about object and event semantics for transfer learning, by proposing a system that uses first-order logic grounded in neural networks to inject symbolic facts into a single agent architecture, and demonstrated in simple experiments that it can learn to utilize both symbolic and image layers in decision-making.
Human ability at solving complex tasks is helped by priors on object and event semantics of their environment. This paper investigates the use of similar prior knowledge for transfer learning in Reinforcement Learning agents. In particular, the paper proposes to use a first-order-logic language grounded in deep neural networks to represent facts about objects and their semantics in the real world. Facts are provided as background knowledge a priori to learning a policy for how to act in the world. The priors are injected with the conventional input in a single agent architecture. As proof-of-concept, the paper tests the system in simple experiments that show the importance of symbolic abstraction and flexible fact derivation. The paper shows that the proposed system can learn to take advantage of both the symbolic layer and the image layer in a single decision selection module.