The Computational Principles of Learning Ability

TL;DR

This paper proposes two fundamental laws defining 'learning ability' in computational terms, aiming to establish a clear, formal understanding of what constitutes a learning model in AI, moving beyond traditional black box tests.

Contribution

It introduces a formal framework with two laws that characterize learning ability and conditions for recognizing a mapping relation as a learning model.

Findings

Defines 'learning ability' through two fundamental laws.

Provides criteria for when a mapping relation qualifies as a learning model.

Bridges the gap between AI performance and theoretical understanding of intelligence.

Abstract

It has been quite a long time since AI researchers in the field of computer science stop talking about simulating human intelligence or trying to explain how brain works. Recently, represented by deep learning techniques, the field of machine learning is experiencing unprecedented prosperity and some applications with near human-level performance bring researchers confidence to imply that their approaches are the promising candidate for understanding the mechanism of human brain. However apart from several ancient philological criteria and some imaginary black box tests (Turing test, Chinese room) there is no computational level explanation, definition or criteria about intelligence or any of its components. Base on the common sense that learning ability is one critical component of intelligence and inspect from the viewpoint of mapping relations, this paper presents two laws which explains what is the "learning ability" as we familiar with and under what conditions a mapping relation can be acknowledged as "Learning Model".