Multiple-Instance Learning by Boosting Infinitely Many Shapelet-based Classifiers

TL;DR

This paper introduces a novel multiple-instance learning framework utilizing infinitely many shapelet-based classifiers, offering improved theoretical guarantees and empirical performance for MIL and time-series classification tasks.

Contribution

It proposes a new formulation of MIL with infinitely many shapelets, along with an efficient algorithm and generalization bounds, advancing beyond heuristic methods.

Findings

Effective for MIL tasks and shapelet learning in time-series classification.

Provides theoretical generalization guarantees for the classifiers.

Empirical results show improved accuracy over existing methods.

Abstract

We propose a new formulation of Multiple-Instance Learning (MIL). In typical MIL settings, a unit of data is given as a set of instances called a bag and the goal is to find a good classifier of bags based on similarity from a single or finitely many "shapelets" (or patterns), where the similarity of the bag from a shapelet is the maximum similarity of instances in the bag. Classifiers based on a single shapelet are not sufficiently strong for certain applications. Additionally, previous work with multiple shapelets has heuristically chosen some of the instances as shapelets with no theoretical guarantee of its generalization ability. Our formulation provides a richer class of the final classifiers based on infinitely many shapelets. We provide an efficient algorithm for the new formulation, in addition to generalization bound. Our empirical study demonstrates that our approach is effective not only for MIL tasks but also for Shapelet Learning for time-series classification.