Deductive Association Networks
This addresses the challenge of high-dimensional thinking in AI, though it appears incremental as it applies a novel method to a standard dataset.
The paper tackles the problem of enabling deductive reasoning in neural networks by introducing Deductive Association Networks (DANs), which combine axioms to infer new propositions, and demonstrates results on the MNIST dataset applied to group theory.
we introduce deductive association networks(DANs), a network that performs deductive reasoning. To have high-dimensional thinking, combining various axioms and putting the results back into another axiom is necessary to produce new relationships and results. For example, it would be given two propositions: "Socrates is a man." and "All men are mortals." and two propositions could be used to infer the new proposition, "Therefore Socrates is mortal.". To evaluate, we used MNIST Dataset, a handwritten numerical image dataset, to apply it to the group theory and show the results of performing deductive learning.