Inference of Abstraction for a Unified Account of Symbolic Reasoning from Data
This work addresses the challenge of integrating symbolic reasoning with probabilistic methods for researchers in AI and cognitive science, though it appears incremental as it builds on existing Bayesian and logical frameworks.
The paper tackles the problem of providing a unified probabilistic account of various types of symbolic reasoning from data, characterizing them using formal logic and consequence relations, with the result of offering new insights for advancing human-like machine intelligence.
Inspired by empirical work in neuroscience for Bayesian approaches to brain function, we give a unified probabilistic account of various types of symbolic reasoning from data. We characterise them in terms of formal logic using the classical consequence relation, an empirical consequence relation, maximal consistent sets, maximal possible sets and maximum likelihood estimation. The theory gives new insights into reasoning towards human-like machine intelligence.