Managing Action Preconditions in Neuro-Symbolic RL: Three Placement Strategies for Embodied Agents

Humans carry behaviour knowledge of how to act in familiar situations into every new task rather than relearning it from scratch. There is no reason a Reinforcement Learning (RL) agent shouldn't do the same: known behaviour patterns need not be learned, only applied. Neuro-symbolic RL bridges prior knowledge and RL by injecting symbolic knowledge alongside a learned policy. The point at which this knowledge is integrated is critical: a poor choice can produce, for instance, hallucinated preconditions, which surface as safety and reliability problems in agents acting in changing environments. We formalise this behavioural knowledge as a precondition Bayesian network (BN) over the agent's \emph{structural actions} - the actions whose legality depends on preconditions, such as picking up a key, grasping a block, toggling a door, or dropping an object. The BN restricts when these actions may fire, and we inject it into the RL loop at three placements: (1) a \emph{symbolic verifier}, consulted only at inference, that fires a structural action once its preconditions hold; (2) a \emph{symbolic enforcer}, active during both training and inference, that governs structural-action use throughout learning; and (3) a \emph{symbolic learner}, which folds the knowledge into the network and learns the restriction and use of structural actions itself. To test the three variants we run experiments on two benchmarks with opposite regimes: one built on long, ordered planning chains, the other on continuous manipulation. We compare against strong baselines on solution quality, sample efficiency, and traceability. The payoff is substantial. On MiniGrid, all three placements improve the \emph{solution quality} over the PPO+RND baseline, the symbolic enforcer leading at $98.2\%$ against the baseline's $88.8\%$. On Fetch, where SAC+HER already solves the task at about $97\%$, solution quality no longer separates the placements. The payoff shifts to \emph{sample efficiency}, with symbolic enforcer and learner reaching that ceiling about $2\times$ sooner. The advantage also holds beyond the embodied benchmarks: on a non-embodied routing task over a real street network, external placement again leads and generalises to held-out instances. Beyond performance, the symbolic verifier and enforcer keep behaviour checkable: a structural action never fires unless its grounded preconditions hold.