Causal Interventional Prediction System for Robust and Explainable Effect Forecasting

TL;DR

This paper introduces a causal interventional prediction system that enhances robustness and explainability in effect forecasting by leveraging causal graphs and variational autoencoders, outperforming existing methods.

Contribution

The work develops a novel causal interventional prediction system (CIPS) that incorporates causal graphs and variational autoencoders for improved robustness and explainability in effect forecasting.

Findings

CIPS outperforms state-of-the-art forecasting methods.

Demonstrates versatility and extensibility in practical applications.

Provides an in-depth causal analysis for effect prediction.

Abstract

Although the widespread use of AI systems in today's world is growing, many current AI systems are found vulnerable due to hidden bias and missing information, especially in the most commonly used forecasting system. In this work, we explore the robustness and explainability of AI-based forecasting systems. We provide an in-depth analysis of the underlying causality involved in the effect prediction task and further establish a causal graph based on treatment, adjustment variable, confounder, and outcome. Correspondingly, we design a causal interventional prediction system (CIPS) based on a variational autoencoder and fully conditional specification of multiple imputations. Extensive results demonstrate the superiority of our system over state-of-the-art methods and show remarkable versatility and extensibility in practice.