Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy

TL;DR

The paper introduces SPRT-TANDEM, a neural network-based sequential testing algorithm that improves early and accurate classification of sequential data by overcoming key limitations of traditional SPRT, especially in correlated data scenarios.

Contribution

It proposes a novel deep learning method, SPRT-TANDEM, that estimates likelihood ratios in correlated data, enhancing early classification accuracy over existing methods.

Findings

Achieves better classification accuracy than baselines.

Uses fewer data samples for decision-making.

Demonstrates effectiveness on multiple video datasets.

Abstract

Classifying sequential data as early and as accurately as possible is a challenging yet critical problem, especially when a sampling cost is high. One algorithm that achieves this goal is the sequential probability ratio test (SPRT), which is known as Bayes-optimal: it can keep the expected number of data samples as small as possible, given the desired error upper-bound. However, the original SPRT makes two critical assumptions that limit its application in real-world scenarios: (i) samples are independently and identically distributed, and (ii) the likelihood of the data being derived from each class can be calculated precisely. Here, we propose the SPRT-TANDEM, a deep neural network-based SPRT algorithm that overcomes the above two obstacles. The SPRT-TANDEM sequentially estimates the log-likelihood ratio of two alternative hypotheses by leveraging a novel Loss function for Log-Likelihood Ratio estimation (LLLR) while allowing correlations up to $N (\in \mathbb{N})$ preceding samples. In tests on one original and two public video databases, Nosaic MNIST, UCF101, and SiW, the SPRT-TANDEM achieves statistically significantly better classification accuracy than other baseline classifiers, with a smaller number of data samples. The code and Nosaic MNIST are publicly available at https://github.com/TaikiMiyagawa/SPRT-TANDEM.