On the Design of PSyKI: a Platform for Symbolic Knowledge Injection into Sub-Symbolic Predictors

A long-standing ambition in artificial intelligence is to integrate predictors’ inductive features (i.e., learning from examples) with deductive capabilities (i.e., drawing inferences from symbolic knowledge). Many methods in the literature support injection of symbolic knowledge into predictors, generally following the purpose of attaining better (i.e., more effective or efficient w.r.t. predictive performance) predictors. However, to the best of our knowledge, running implementations of these algorithms are currently either proof of concepts or unavailable in most cases. Moreover, a unified, coherent software framework supporting them as well as their interchange, comparison, and exploitation in arbitrary ML workflows is currently missing. Accordingly, in this paper we present the design of PSyKI, a platform providing general-purpose support to symbolic knowledge injection into predictors via different algorithms. In particular, we discuss the overall architecture, and the many components/functionalities of PSyKI, invidually—providing examples as well. We finally demonstrate the versatility of our approach by exemplifying two custom injection algorithms in a toy scenario: Poker Hands classification.