Causal Sensitivity Identification using Generative Learning
This work addresses the need for causal inference in prediction tasks, such as spatiotemporal trajectory recommendations, but is incremental as it builds on existing generative models like CVAE.
The paper tackles the problem of identifying causal impacts for prediction tasks by proposing a generative method using Conditional Variational Autoencoders (CVAE) to reduce confounding bias and improve predictive performance, as validated on the GeoLife dataset and Asia Bayesian network benchmark.
In this work, we propose a novel generative method to identify the causal impact and apply it to prediction tasks. We conduct causal impact analysis using interventional and counterfactual perspectives. First, applying interventions, we identify features that have a causal influence on the predicted outcome, which we refer to as causally sensitive features, and second, applying counterfactuals, we evaluate how changes in the cause affect the effect. Our method exploits the Conditional Variational Autoencoder (CVAE) to identify the causal impact and serve as a generative predictor. We are able to reduce confounding bias by identifying causally sensitive features. We demonstrate the effectiveness of our method by recommending the most likely locations a user will visit next in their spatiotemporal trajectory influenced by the causal relationships among various features. Experiments on the large-scale GeoLife [Zheng et al., 2010] dataset and the benchmark Asia Bayesian network validate the ability of our method to identify causal impact and improve predictive performance.