Predicate Invention by Learning From Failures
This addresses a crucial but notoriously difficult challenge in ILP for advancing AI towards human-level concept discovery, though it appears incremental as it builds on existing ILP and answer set programming methods.
The paper tackles the problem of predicate invention (PI) in inductive logic programming (ILP) by introducing POPPI, a system that formulates PI as an answer set programming problem, resulting in drastically improved learning performance and substantial outperformance of existing ILP systems.
Discovering novel high-level concepts is one of the most important steps needed for human-level AI. In inductive logic programming (ILP), discovering novel high-level concepts is known as predicate invention (PI). Although seen as crucial since the founding of ILP, PI is notoriously difficult and most ILP systems do not support it. In this paper, we introduce POPPI, an ILP system that formulates the PI problem as an answer set programming problem. Our experiments show that (i) PI can drastically improve learning performance when useful, (ii) PI is not too costly when unnecessary, and (iii) POPPI can substantially outperform existing ILP systems.