Fast classification using sparse decision DAGs

TL;DR

This paper introduces a novel algorithm for constructing sparse decision DAGs from base classifiers, optimizing the accuracy-speed trade-off, and demonstrating superior performance in object detection and web ranking tasks.

Contribution

The paper presents a new method to build sparse decision DAGs using a Markov decision process, enabling data-dependent classifier selection and improved efficiency over cascade methods.

Findings

Outperforms state-of-the-art cascade detectors on object detection benchmarks.

Significantly improves decision speed in web ranking without loss of performance.

Applicable to multi-class classification, unlike traditional cascades.

Abstract

In this paper we propose an algorithm that builds sparse decision DAGs (directed acyclic graphs) from a list of base classifiers provided by an external learning method such as AdaBoost. The basic idea is to cast the DAG design task as a Markov decision process. Each instance can decide to use or to skip each base classifier, based on the current state of the classifier being built. The result is a sparse decision DAG where the base classifiers are selected in a data-dependent way. The method has a single hyperparameter with a clear semantics of controlling the accuracy/speed trade-off. The algorithm is competitive with state-of-the-art cascade detectors on three object-detection benchmarks, and it clearly outperforms them when there is a small number of base classifiers. Unlike cascades, it is also readily applicable for multi-class classification. Using the multi-class setup, we show on a benchmark web page ranking data set that we can significantly improve the decision speed without harming the performance of the ranker.