A neuro-symbolic framework for sequence classification with relational and temporal knowledge

One of the goals of neuro-symbolic artificial intel- ligence is to exploit background knowledge to im- prove the performance of learning tasks. However, most of the existing frameworks focus on the sim- plified scenario where knowledge does not change over time and does not cover the temporal dimen- sion. In this work we consider the much more chal- lenging problem of knowledge-driven sequence classification where different portions of knowl- edge must be employed at different timesteps, and temporal relations are available. Our experimental evaluation compares multi-stage neuro-symbolic and neural-only architectures, and it is conducted on a newly-introduced benchmarking framework. Results demonstrate the challenging nature of this novel setting, and also highlight under-explored shortcomings of neuro-symbolic methods, repre- senting a precious reference for future research.