Neural Reward Machines
This addresses the challenge of incorporating symbolic knowledge in continuous or non-symbolic RL domains for AI researchers, though it is incremental as it builds on existing automata-based approaches.
The authors tackled the problem of non-Markovian reinforcement learning in non-symbolic environments by introducing Neural Reward Machines, a neurosymbolic framework that combines RL with semisupervised symbol grounding, outperforming Deep RL methods and improving grounding efficiency by a factor of 10^3.
Non-markovian Reinforcement Learning (RL) tasks are very hard to solve, because agents must consider the entire history of state-action pairs to act rationally in the environment. Most works use symbolic formalisms (as Linear Temporal Logic or automata) to specify the temporally-extended task. These approaches only work in finite and discrete state environments or continuous problems for which a mapping between the raw state and a symbolic interpretation is known as a symbol grounding (SG) function. Here, we define Neural Reward Machines (NRM), an automata-based neurosymbolic framework that can be used for both reasoning and learning in non-symbolic non-markovian RL domains, which is based on the probabilistic relaxation of Moore Machines. We combine RL with semisupervised symbol grounding (SSSG) and we show that NRMs can exploit high-level symbolic knowledge in non-symbolic environments without any knowledge of the SG function, outperforming Deep RL methods which cannot incorporate prior knowledge. Moreover, we advance the research in SSSG, proposing an algorithm for analysing the groundability of temporal specifications, which is more efficient than baseline techniques of a factor $10^3$.