Backward Training Explained
Backward training operates through two main functions, "or" and "and," which work recursively to prove goals by breaking them down into sub-goals. The process relies heavily on unification, where symbols are aligned to facilitate the comparison of goals with rules. This method allows for a systematic approach to knowledge verification, enhancing the understanding of symbolic reasoning in AI.In this clip
From this podcast

NLP Highlights
19 - End-to-end Differentiable Proving, with Tim Rocktäschel
Related Questions