Abstraction Learning
This work addresses the fundamental problem of bridging AI and human intelligence, but it appears incremental as it builds on prior concepts of abstraction in AI.
The paper tackles the gap between artificial and human intelligence by proposing abstraction learning as a solution, introducing the ONE framework that addresses representation, objective function, and learning algorithm challenges, and demonstrates on MNIST that it achieves low energy consumption, knowledge sharing, and lifelong learning.
There has been a gap between artificial intelligence and human intelligence. In this paper, we identify three key elements forming human intelligence, and suggest that abstraction learning combines these elements and is thus a way to bridge the gap. Prior researches in artificial intelligence either specify abstraction by human experts, or take abstraction as a qualitative explanation for the model. This paper aims to learn abstraction directly. We tackle three main challenges: representation, objective function, and learning algorithm. Specifically, we propose a partition structure that contains pre-allocated abstraction neurons; we formulate abstraction learning as a constrained optimization problem, which integrates abstraction properties; we develop a network evolution algorithm to solve this problem. This complete framework is named ONE (Optimization via Network Evolution). In our experiments on MNIST, ONE shows elementary human-like intelligence, including low energy consumption, knowledge sharing, and lifelong learning.