Analogical Reasoning Within a Conceptual Hyperspace
This work addresses the challenge of implementing analogical reasoning for AI systems, though it is incremental as it builds on existing theories with preliminary results.
The authors tackled the problem of operationalizing analogical inference by combining hyperdimensional computing with Conceptual Spaces Theory, resulting in a concrete architecture that demonstrated category-based and property-based analogical reasoning in a toy domain.
We propose an approach to analogical inference that marries the neuro-symbolic computational power of complex-sampled hyperdimensional computing (HDC) with Conceptual Spaces Theory (CST), a promising theory of semantic meaning. CST sketches, at an abstract level, approaches to analogical inference that go beyond the standard predicate-based structure mapping theories. But it does not describe how such an approach can be operationalized. We propose a concrete HDC-based architecture that computes several types of analogy classified by CST. We present preliminary proof-of-concept experimental results within a toy domain and describe how it can perform category-based and property-based analogical reasoning.