CausalTAD: Causal Implicit Generative Model for Debiased Online Trajectory Anomaly Detection

TL;DR

CausalTAD introduces a causal inference-based generative model to improve trajectory anomaly detection by removing confounding biases, enhancing performance on both trained and out-of-distribution data.

Contribution

The paper proposes CausalTAD, a novel causal implicit generative model that uses do-calculus to address confounding bias in trajectory anomaly detection.

Findings

Achieves 2.1% to 5.7% improvement on trained trajectories.

Achieves 10.6% to 32.7% improvement on out-of-distribution data.

Effectively eliminates confounding bias from road network preferences.

Abstract

Trajectory anomaly detection, aiming to estimate the anomaly risk of trajectories given the Source-Destination (SD) pairs, has become a critical problem for many real-world applications. Existing solutions directly train a generative model for observed trajectories and calculate the conditional generative probability $P({T}|{C})$ as the anomaly risk, where ${T}$ and ${C}$ represent the trajectory and SD pair respectively. However, we argue that the observed trajectories are confounded by road network preference which is a common cause of both SD distribution and trajectories. Existing methods ignore this issue limiting their generalization ability on out-of-distribution trajectories. In this paper, we define the debiased trajectory anomaly detection problem and propose a causal implicit generative model, namely CausalTAD, to solve it. CausalTAD adopts do-calculus to eliminate the confounding bias of road network preference and estimates $P({T}|do({C}))$ as the anomaly criterion. Extensive experiments show that CausalTAD can not only achieve superior performance on trained trajectories but also generally improve the performance of out-of-distribution data, with improvements of $2.1\% \sim 5.7\%$ and $10.6\% \sim 32.7\%$ respectively.