On the Solvability of Inductive Problems: A Study in Epistemic Topology
This work addresses foundational issues in epistemology and AI for researchers in logic and formal learning theory, offering theoretical insights into universal solvability.
The paper tackles the problem of inductive problem-solving and learning by doxastic agents, providing topological characterizations of solvability and learnability and proving that AGM-style belief revision is universal, meaning every solvable problem can be solved by AGM conditioning.
We investigate the issues of inductive problem-solving and learning by doxastic agents. We provide topological characterizations of solvability and learnability, and we use them to prove that AGM-style belief revision is "universal", i.e., that every solvable problem is solvable by AGM conditioning.