Specific-to-General Learning for Temporal Events with Application to Learning Event Definitions from Video

TL;DR

This paper introduces a novel supervised specific-to-general learning algorithm for temporal event definitions in videos, using a simple propositional temporal logic called AMA, and demonstrates its effectiveness in learning from video sequences.

Contribution

The paper presents a new specific-to-general learning algorithm for temporal event definitions using AMA logic, with complexity analysis and application to video event learning.

Findings

The AMA language effectively represents many events.

The learning algorithm produces competitive event definitions from videos.

The syntactic subsumption test improves computational efficiency.

Abstract

We develop, analyze, and evaluate a novel, supervised, specific-to-general learner for a simple temporal logic and use the resulting algorithm to learn visual event definitions from video sequences. First, we introduce a simple, propositional, temporal, event-description language called AMA that is sufficiently expressive to represent many events yet sufficiently restrictive to support learning. We then give algorithms, along with lower and upper complexity bounds, for the subsumption and generalization problems for AMA formulas. We present a positive-examples--only specific-to-general learning method based on these algorithms. We also present a polynomial-time--computable ``syntactic'' subsumption test that implies semantic subsumption without being equivalent to it. A generalization algorithm based on syntactic subsumption can be used in place of semantic generalization to improve the asymptotic complexity of the resulting learning algorithm. Finally, we apply this algorithm to the task of learning relational event definitions from video and show that it yields definitions that are competitive with hand-coded ones.