Abstraction and Analogy-Making in Artificial Intelligence
It addresses the problem of enhancing AI's learning and reasoning capabilities for researchers, but is incremental as it reviews existing methods and suggests future directions.
The paper reviews approaches for enabling AI systems to achieve human-like conceptual abstraction and analogy-making, concluding with proposals for challenge tasks and evaluation measures to facilitate quantifiable progress.
Conceptual abstraction and analogy-making are key abilities underlying humans' abilities to learn, reason, and robustly adapt their knowledge to new domains. Despite of a long history of research on constructing AI systems with these abilities, no current AI system is anywhere close to a capability of forming humanlike abstractions or analogies. This paper reviews the advantages and limitations of several approaches toward this goal, including symbolic methods, deep learning, and probabilistic program induction. The paper concludes with several proposals for designing challenge tasks and evaluation measures in order to make quantifiable and generalizable progress in this area.