An Approximate Bayesian Long Short-Term Memory Algorithm for Outlier Detection

TL;DR

This paper introduces an approximate Bayesian LSTM algorithm utilizing Ensemble Kalman Filter for outlier detection, addressing overfitting and uncertainty estimation issues in neural networks.

Contribution

It proposes a scalable Bayesian LSTM method with optimized noise covariance estimation for improved outlier detection in real-world data.

Findings

Effective outlier detection on Twitter data

Scalable Bayesian weight uncertainty estimation

Enhanced uncertainty quantification in LSTM models

Abstract

Long Short-Term Memory networks trained with gradient descent and back-propagation have received great success in various applications. However, point estimation of the weights of the networks is prone to over-fitting problems and lacks important uncertainty information associated with the estimation. However, exact Bayesian neural network methods are intractable and non-applicable for real-world applications. In this study, we propose an approximate estimation of the weights uncertainty using Ensemble Kalman Filter, which is easily scalable to a large number of weights. Furthermore, we optimize the covariance of the noise distribution in the ensemble update step using maximum likelihood estimation. To assess the proposed algorithm, we apply it to outlier detection in five real-world events retrieved from the Twitter platform.