Induction, of and by Probability

TL;DR

This paper discusses probabilistic methods for learning systems, focusing on heuristic search, where probabilities guide task performance and learning, enabling noise management and incremental improvement.

Contribution

It introduces probabilistic principles for generalization learning, demonstrating their application in heuristic search with noise management and incremental updates.

Findings

Probabilistic normalization improves learning accuracy

Incremental probability revision enhances generalization

Effective noise management in heuristic search

Abstract

This paper examines some methods and ideas underlying the author's successful probabilistic learning systems(PLS), which have proven uniquely effective and efficient in generalization learning or induction. While the emerging principles are generally applicable, this paper illustrates them in heuristic search, which demands noise management and incremental learning. In our approach, both task performance and learning are guided by probability. Probabilities are incrementally normalized and revised, and their errors are located and corrected.