CausalTAD: Causal Implicit Generative Model for Debiased Online Trajectory Anomaly Detection
It solves a debiased anomaly detection problem for real-world applications like traffic monitoring, but it is incremental as it builds on existing generative models with causal adjustments.
The paper tackles trajectory anomaly detection by addressing confounding bias from road network preferences, proposing CausalTAD to estimate debiased anomaly risk, resulting in performance improvements of 2.1% to 5.7% on trained trajectories and 10.6% to 32.7% on out-of-distribution data.
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.