Learning where and when to reason in neuro-symbolic inference

The imposition of hard constraints on the output of neural networks is a highly desirable capability, as it instills confidence in AI by ensuring that neural network predictions adhere to domain expertise. This area has received significant attention recently, however, current methods typically enforce constraints in a "weak" form during training, with no guarantees at inference, and do not provide a general framework for different tasks/constraint types. We approach this open problem from a neuro-symbolic perspective. Our method enhances a conventional neural predictor with a reasoning module that can correct predictions errors and a neural attention module that learns to focus the reasoning effort on potential prediction errors while leaving other outputs unchanged. This framework provides a balance between the efficiency of unconstrained neural inference and the high cost of exhaustive reasoning during inference.